video_id
stringlengths 8
11
| video
stringlengths 22
25
| conversations
listlengths 2
2
|
|---|---|---|
seed42_2688
|
videos/HRA_part0/2688.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An experienced engineer, clad in a yellow hard hat, safety glasses, and a blue uniform, stands amidst the towering structures and pipelines of an oil refinery complex. He is seen wiping his hands on a cloth, perhaps after completing a task, with the industrial landscape of the facility stretching behind him. The scene captures the essence of hard work and maintenance in the heart of the oil industry.]"
},
{
"from": "gpt",
"value": "Bad. The engineer's hand wiping in the video is very unsmooth and the movements are uncoordinated."
}
] |
seed46_1544
|
videos/HRA_part2/1544.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A worker, clad in a brown uniform, diligently wipes the glass enclosure of a wildlife exhibit, where a graceful gazelle is seen resting calmly inside. The enclosure is meticulously clean, save for a few lingering spots that the worker is addressing. The gazelle, displaying a relaxed demeanor, is positioned near the enclosure's entrance, its body language indicating a sense of security and comfort within its habitat. The scene captures the routine maintenance of the animal's living space, highlighting the careful interaction between the human caretaker and the wild creature in a shared environment.]"
},
{
"from": "gpt",
"value": "Bad. The human hand appears deformed and mutated."
}
] |
seed44_291
|
videos/HRA_part1/291.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A casually dressed doctor, wearing a white button-down shirt and khaki shorts, leaps into the air on a pristine, sandy beach, with crystal-clear water visible in the background. The doctor, who has brown hair and glasses, is captured mid-jump, displaying a sense of freedom and joy. The scene is set against a stunning backdrop of palm trees and a clear blue sky, creating a perfect blend of professional and leisure elements. The doctor's relaxed attire and the idyllic beach setting convey a sense of balance between work and personal life.]"
},
{
"from": "gpt",
"value": "Bad. The doctor's face in the video is generated blurry, and the limbs are deformed."
}
] |
seed44_1628
|
videos/HRA_part1/1628.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A quirky writer, adorned in a bat costume complete with a black cape and headpiece, revels in a solo dance within a dimly lit room. Surrounded by scattered papers and an old typewriter, the scene opens with the writer seated pensively at a desk, illuminated by a single lamp. Suddenly, they leap up, transforming into a dancing figure, their movements wild and uninhibited, expressing the creative energy coursing through them. The bat-themed outfit adds a whimsical touch to the passionate dance, embodying the fusion of writing and movement in a unique celebration of artistic inspiration.]"
},
{
"from": "gpt",
"value": "Bad. The character's face and limbs in the video are distorted."
}
] |
seed46_1637
|
videos/HRA_part2/1637.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A golden retriever, framed against a vibrant green meadow under a clear blue sky, experiences a unique perception of time as it runs. Initially, the dog is seen standing still, looking forward with anticipation. It then starts running, its legs blurring as if time has accelerated for the viewer. Next, the dog leaps into the air, seemingly frozen in mid-air, giving the impression of time slowing down. Finally, the dog bounds across the field, its motion smooth and unhurried, conveying a sense of normal time flow, all while the lush green landscape and clear blue sky provide a serene backdrop.]"
},
{
"from": "gpt",
"value": "Bad. The dog's limbs and paws are missing in the video, and the tail has a double image."
}
] |
seed44_1491
|
videos/HRA_part1/1491.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A college-aged student, dressed in a gray hoodie and blue jeans, diligently sweeps the wooden planks of a front porch bathed in natural sunlight, the fibers of the broom bristles catching the light as they sweep away dust and leaves. The porch railing, painted white, contrasts against the weathered wood, and the lush greenery of the surrounding landscape provides a vibrant backdrop. The young man pauses for a moment, leaning on the broom, revealing a tidy, swept area underfoot, before continuing his task with a focused expression, the sunlight casting a warm glow over the entire scene.]"
},
{
"from": "gpt",
"value": "Good. Basically no problem."
}
] |
seed44_2882
|
videos/HRA_part1/2882.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A powerful lion, defying expectations, swims through a tranquil, crystal-clear lake, its muscular body moving gracefully beneath the water's surface. The big cat's tawny coat glistens as it propels itself forward with strong, deliberate strokes. Its piercing gaze remains fixed ahead, showcasing its innate strength and adaptability in an unexpected aquatic environment. The scene is set against a lush, verdant landscape, highlighting the unique and captivating behavior of this typically terrestrial predator.]"
},
{
"from": "gpt",
"value": "Good. Basically no problem."
}
] |
seed42_159
|
videos/HRA_part0/159.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A large, male gorilla stands in a lush, green forest, engaging in a series of comical dance moves. He begins by swaying side to side, his massive body moving in time with an unseen rhythm. Suddenly, he breaks into a playful jig, his arms flailing wildly as he hops from foot to foot. The gorilla then attempts a spin, his wide grin revealing a mouthful of sharp teeth. Throughout the performance, the gorilla remains in the same spot, his dance a delightful contrast to the stillness of the surrounding environment. Despite his size and strength, the gorilla's movements are surprisingly graceful, his joyous expression captivating as he continues to dance in the heart of the forest.]"
},
{
"from": "gpt",
"value": "Bad. The gorilla's dance in the video is not smooth, and the movements are uncoordinated."
}
] |
seed42_1315
|
videos/HRA_part0/1315.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A student, dressed in a yellow t-shirt and blue jeans, is playfully chasing a juvenile rhinoceros across a vast, grassy savannah under a bright blue sky. The action is captured from multiple angles, first showing the student's enthusiastic pursuit, followed by a shot of the rhino's surprised reaction as it turns its head back. Next, the scene focuses on the student's determined facial expression, highlighting the joy and energy of the chase. Finally, a side view captures the dynamic duo running in unison, the student's arms waving wildly and the rhinoceros' ears flapping, both enjoying the spontaneous adventure in the expansive landscape.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed42_2667
|
videos/HRA_part0/2667.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A musician, wearing a black jacket and blue jeans, gently caresses the head of a bison, which is lying down on a grassy field under a cloudy sky. The musician's left hand rests on the bison's snout, while the animal's eye is closed in apparent comfort. Both figures are set against a backdrop of green fields and distant mountains, with the bison's body partially obscured by the tall grass. The musician's right hand holds a harmonica, which hangs loosely by its strap, adding a musical element to the scene of connection between human and animal.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed44_1628
|
videos/HRA_part1/1628.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A quirky writer, adorned in a bat costume complete with a black cape and headpiece, revels in a solo dance within a dimly lit room. Surrounded by scattered papers and an old typewriter, the scene opens with the writer seated pensively at a desk, illuminated by a single lamp. Suddenly, they leap up, transforming into a dancing figure, their movements wild and uninhibited, expressing the creative energy coursing through them. The bat-themed outfit adds a whimsical touch to the passionate dance, embodying the fusion of writing and movement in a unique celebration of artistic inspiration.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed46_2414
|
videos/HRA_part2/2414.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a lush, verdant orchard filled with apple trees heavy with ripe fruit, a large brown bear stands on its hind legs, its shaggy fur glistening in the dappled sunlight. It reaches up with a massive paw, gently swiping at the low-hanging branches laden with apples. The bear then lowers itself to all fours, picking up an apple it has dislodged, and wipes it carefully on its thick fur before taking a hearty bite, savoring the sweet, juicy flavor amidst the peaceful surroundings of the orchard.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed46_2823
|
videos/HRA_part2/2823.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A magnificent bird, identified as a golden eagle, soars gracefully against a clear blue sky, its wings spread wide to catch the wind. The eagle's piercing yellow eyes and powerful beak are visible as it glides effortlessly, surveying the vast landscape below. Its majestic plumage, featuring shades of brown and gold, contrasts beautifully with the azure heavens. In another shot, the eagle is seen perched on a rocky outcropping, its sharp talons gripping tightly as it turns its head, alert and watchful, against a backdrop of rolling hills and distant mountains.]"
},
{
"from": "gpt",
"value": "Good. No obvious issues."
}
] |
seed44_2925
|
videos/HRA_part1/2925.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A charismatic male musician, dressed in a black leather jacket, dark jeans, and sunglasses, walks purposefully down a busy city street, the iconic Flatiron Building visible in the background. Suddenly, the scene shifts to depict him walking at an extraordinary pace, a visual effect suggesting the concept of light-speed travel. He continues his brisk walk, now surrounded by blurs of color and light, evoking the sensation of moving at incredible velocities. Throughout this dynamic sequence, he maintains his cool demeanor, his outfit unchanged, embodying the fusion of urban style and science fiction themes.]"
},
{
"from": "gpt",
"value": "Bad. The male musician's face generation in the video is uncoordinated, and his fingers are deformed."
}
] |
seed42_269
|
videos/HRA_part0/269.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a serene, misty forest setting, a majestic bull moose with a full rack of antlers tenderly nuzzles a smaller female moose, both enveloped in the early morning light filtering through the trees. Their interaction is gentle and intimate, showcasing the softer side of these typically solitary creatures. The male's large, furry body and impressive antlers contrast with the delicate nature of their caress, highlighting the bond between them. The scene is set against a backdrop of lush vegetation and tall trees, creating a peaceful and private moment in the wild.]"
},
{
"from": "gpt",
"value": "Normal. The antlers of the red moose in the video are over-generated, one ear is missing, the facial features of the fawn are not generated, and the limbs are missing."
}
] |
seed46_2710
|
videos/HRA_part2/2710.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An actor, resembling Colin Farrell, is seen walking through a dense forest with tall trees, wearing a brown jacket, jeans, and boots. Suddenly, the scene changes to reveal a prehistoric setting, where the actor now finds himself casually strolling beside a woolly mammoth. Both the actor and the mammoth continue their leisurely walk together, with the mammoth's massive frame towering over the actor, who appears unfazed by the extraordinary situation. They meander through the ancient landscape, surrounded by lush vegetation and underbrush, creating a surreal juxtaposition of eras and species.]"
},
{
"from": "gpt",
"value": "Normal. The person's face, hands, and arms are distorted in the video, and the mammoth's head is also twisted."
}
] |
seed42_1304
|
videos/HRA_part0/1304.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A muscular man with short brown hair, wearing a black wetsuit, swims underwater in a clear, blue pool, interacting closely with a live cobra that moves alongside him. The snake, its hood slightly spread, appears calm in the water, its body parallel to the man's outstretched arm. Sunlight filters through the water, casting dappled light on the man's face as he maintains eye contact with the snake, both creatures moving gracefully through the aquatic environment. The scene is set in an outdoor pool, surrounded by lush greenery, emphasizing the natural setting and the unique interaction between human and wildlife.]"
},
{
"from": "gpt",
"value": "The man's swimming action is basically normal, but the cobra's swimming action is unsmooth."
}
] |
seed42_1220
|
videos/HRA_part0/1220.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished director, donning a black baseball cap, black shirt, khaki pants, and black shoes, is shown repairing a vintage bicycle against a backdrop of lush greenery and sunlight filtering through the trees. He is engaged in fixing the bike's chain, showcasing his mechanical skills. In another shot, he carefully inspects the bicycle's frame, his face expressing concentration and expertise. He then tightens a bolt on the bike's wheel, demonstrating his proficiency in bike maintenance. Lastly, he adjusts the bicycle's seat, ensuring it is perfectly aligned, all while surrounded by the serene beauty of nature.]"
},
{
"from": "gpt",
"value": "Normal. The director's movements to repair the bicycle in the video are not smooth and coordinated."
}
] |
seed46_1452
|
videos/HRA_part2/1452.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vibrant, imaginative scene, a woman with a pixie haircut, dressed in a NASA astronaut suit, interacts playfully with a capuchin monkey wearing a matching miniature spacesuit, both standing inside a futuristic space education center. The room is filled with interactive displays and holographic projections, creating an immersive learning environment. Moments later, they are seen together in an indoor pool, designed to simulate zero-gravity conditions, where the woman and the monkey float effortlessly in their spacesuits, highlighting the unique experience of swimming in a space-themed educational facility.]"
},
{
"from": "gpt",
"value": "Bad. The scene in the pool is missing the woman."
}
] |
seed44_2545
|
videos/HRA_part1/2545.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a natural, woody setting, a man with a beard, wearing a brown jacket, sits on the ground, gently caressing a gray wolf that lies calmly beside him, its head resting on the man's lap. The wolf appears relaxed and trusting, as if sharing a moment of mutual understanding with the man, who looks on with a soft, compassionate gaze, the lush forest and tall trees providing a serene backdrop to their connection.]"
},
{
"from": "gpt",
"value": "Normal. The man's hand and the gray wolf's head generation have slight issues."
}
] |
seed44_2039
|
videos/HRA_part1/2039.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal cosmic setting, a majestic cheetah, adorned with a golden collar and walking on zero-gravity platforms, traverses through an infinite space library. Surrounded by floating bookshelves filled with ancient tomes, the cheetah moves gracefully, its gaze focused on a distant celestial body. The library, illuminated by ethereal starlight, showcases an array of knowledge from across the universe, with the cheetah embodying wisdom and power as it explores this otherworldly domain.]"
},
{
"from": "gpt",
"value": "Bad. The golden necklace mentioned in the caption does not appear in the video."
}
] |
seed42_2279
|
videos/HRA_part0/2279.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A farmer, clad in a brown jacket, beige trousers, and a wide-brimmed hat, dances jovially in a mountainous landscape, the distant peaks shrouded in mist. A deer, with a full rack of antlers and a curious gaze, watches him from the edge of a lush forest. The farmer's movements are lively and carefree, his arms raised in the air, as if celebrating the beauty of nature that surrounds him. The deer, sensing no threat, cautiously approaches, its ears perked, adding a touch of whimsy to the scene. The mountains loom majestically in the background, their slopes dotted with trees and patches of greenery, providing a stunning natural backdrop to this unexpected dance.]"
},
{
"from": "gpt",
"value": "Good. No obvious issues."
}
] |
seed42_2701
|
videos/HRA_part0/2701.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vividly surreal scene, a mongoose, typically a small terrestrial carnivore, is depicted engaging in an extraordinary activity. It is shown riding a horse, specifically a Mars horse, which could imply a horse adapted to the conditions of Mars or having Martian features, across a barren, rocky landscape reminiscent of the Martian surface. The mongoose, now a rider, steers the horse with determination as they race through the rugged terrain. The action intensifies as they begin to run at full speed, the horse's hooves kicking up red dust, while the mongoose holds on tightly, its small size contrasting with the grandeur of the Martian environment. This fantastical scenario combines elements of adventure, science fiction, and nature in a captivating visual narrative.]"
},
{
"from": "gpt",
"value": "Bad. The generation of the mongoose and horse's limbs in the video has problems."
}
] |
seed46_1783
|
videos/HRA_part2/1783.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An actor, costumed in a vibrant green shirt and brown pants, strolls alongside a life-sized, realistic hippopotamus puppet, which is operated by two puppeteers dressed in all-black attire. The trio walks across a grassy field under a clear blue sky, creating a whimsical and unexpected scene. The actor, with a bemused smile, interacts with the puppet, which responds with lifelike movements and expressions, adding to the surreal atmosphere of the outdoor setting.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed46_2770
|
videos/HRA_part2/2770.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An ibex, distinguished by its curved horns and brown coat, walks steadily across a rocky terrain, with a clear blue sky and distant mountains providing a scenic backdrop. The ibex navigates the rugged landscape with ease, its hooves firmly gripping the stone surface. It pauses momentarily, its head held high, surveying its surroundings before continuing its journey, moving gracefully among the boulders and cliffs. Throughout the scene, the ibex remains the central focus, its presence commanding attention against the stark beauty of the alpine environment.]"
},
{
"from": "gpt",
"value": "Normal. The video shows the walking of a ibex, the movements are basically normal but there is some lack of coordination."
}
] |
seed44_1779
|
videos/HRA_part1/1779.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A lone jackal walks through ancient ruins, its sandy fur blending into the weathered stone walls of a historical site bathed in warm sunlight. The camera captures the jackal from various angles, showcasing its cautious exploration of the site, which features tall columns and ornate carvings, hinting at the grandeur of a long-lost civilization. The jackal pauses, its ears perked, scanning the surroundings before continuing its journey through the silent, time-worn corridors, the texture of the stones and the play of light adding depth to the scene.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed46_32
|
videos/HRA_part2/32.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A well-dressed lawyer, complete with a tailored suit, tie, and leather briefcase, finds himself in an unexpected scenario: cleaning windows. With a microfiber cloth in hand, he meticulously wipes a large, clear window pane, revealing a cityscape beyond. The scene is set in a spacious office with modern decor, where natural light illuminates the room, highlighting the lawyer's dedication to detail in an unconventional task. Despite his professional attire, he approaches the chore with the same seriousness and precision as his legal work.]"
},
{
"from": "gpt",
"value": "Normal. The lawyer's face in the video is generated blurry, and the fingers are deformed."
}
] |
seed46_3191
|
videos/HRA_part2/3191.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A prehistoric mammoth, its woolly coat dark and shaggy, is seen wading through a cold, icy lake during the last Ice Age. Its large body creates ripples in the water, and it moves with a surprising grace for its size. The mammoth uses its tusks to navigate through the submerged branches and logs, demonstrating its adaptability to aquatic environments. Despite the freezing temperature, the mammoth appears comfortable, showcasing its ability to thrive in harsh conditions. The backdrop is a lush, green landscape with distant mountains, highlighting the diversity of habitats these ancient creatures inhabited.]"
},
{
"from": "gpt",
"value": "Bad. The mammoth's head and limbs in the video are generated inconsistently."
}
] |
seed44_1831
|
videos/HRA_part1/1831.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a doctor, wearing a white lab coat and blue gloves, is seen wiping the nose of a friendly coyote who is sitting calmly on a checkered medical table. The room is brightly lit, with medical supplies neatly arranged on shelves in the background. The doctor's compassionate expression and the coyote's trusting demeanor create a heartwarming moment, highlighting an unusual yet tender interaction between human and wild animal in a clinical setting.]"
},
{
"from": "gpt",
"value": "Normal. The movements of the doctor and the coyote in the video are basically normal, with slight disharmony."
}
] |
seed42_2748
|
videos/HRA_part0/2748.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a rustic vineyard setting, a young female cashier, dressed in a red apron over a white blouse, diligently wipes down the wooden counter of the quaint tasting room. Sunlight streams in through large windows, highlighting the rich textures of the wood and the lush greenery outside. The atmosphere is peaceful and inviting, with the sound of gentle music playing in the background and the distant hum of vineyard activities. The cashier's focused demeanor and meticulous cleaning convey a sense of pride in her work and the establishment, creating a welcoming environment for visitors.]"
},
{
"from": "gpt",
"value": "Bad. The person's face, hands, and arms are distorted in the video, as well as the item they are holding."
}
] |
seed42_2650
|
videos/HRA_part0/2650.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a sophisticated office setting, a professional lawyer, dressed in a navy suit, white shirt, and striped tie, sits at a cluttered desk adorned with legal documents and a laptop. He pauses from his work to look up, captivated by a stunning holographic display of an exoplanet orbiting its star, projected from his device. The lawyer then wipes away the hologram with a sweeping hand gesture, returning his attention to the challenging legal briefs spread out before him, highlighting the contrast between his daily duties and the captivating world of astronomical discoveries.]"
},
{
"from": "gpt",
"value": "Bad. The person's hand is distorted in the video."
}
] |
seed42_2192
|
videos/HRA_part0/2192.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged writer, dressed in a gray T-shirt, blue jeans, and brown shoes, organizes his garage with determination. He begins by sorting through boxes, revealing a cluttered space filled with various items, including bicycles, tools, and old books. As he works, he meticulously arranges the contents, creating order from chaos. The garage walls, adorned with shelves and pegboards, display an array of neatly organized objects, showcasing the writer's commitment to tidiness. Throughout the process, the natural light filtering through the garage windows highlights his efforts, emphasizing the transformation of the space.]"
},
{
"from": "gpt",
"value": "Bad. The person's arm is twisted, the hand is twisted, the face is distorted, the leg is twisted, and the object in the hand is twisted."
}
] |
seed44_2242
|
videos/HRA_part1/2242.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man wearing a blue shirt, tan shorts, and brown shoes sweeps his wooden porch with a broom, against the backdrop of a sunny day and a lush green forest. He pauses for a moment to look into the distance, where a lake can be seen shimmering under the sunlight. The man resumes sweeping, revealing a clean, well-maintained porch. Suddenly, a playful dog rushes towards him, and the man stops to pat the dog affectionately, before continuing his task. The scene concludes with the man leaning against the porch railing, enjoying the peaceful surroundings and the company of his furry friend.]"
},
{
"from": "gpt",
"value": "Bad. The man's limbs are missing in the video, his face is generated blurry, and the cleaning tools and dog are deformed."
}
] |
seed44_2065
|
videos/HRA_part1/2065.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a warm, sandy landscape under a clear blue sky, a woman wearing a flowing, cream-colored robe gently caresses a kneeling camel adorned with ornate saddles and blankets. The scene captures the tender bond between the woman and the camel, with her hand softly stroking its head. The woman's attire complements the desert environment, and her gentle touch conveys a sense of care and connection. Nearby, more camels stand, their presence adding to the serene atmosphere of the desert setting.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed44_1303
|
videos/HRA_part1/1303.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A North American porcupine, distinguished by its brown and white quills, runs through a dense forest environment, navigating between moss-covered trees. The porcupine, with its quills standing out prominently, dashes across the screen from right to left, showcasing its agility. Its facial features, including small eyes and a pointed nose, are clearly visible as it moves quickly, demonstrating the animal's unexpected speed and grace in its natural habitat.]"
},
{
"from": "gpt",
"value": "Normal. The porcupine's nose in the video is problematic, there is a clipping issue when the porcupine goes through the tree, and the porcupine's feet are not generated."
}
] |
seed44_2941
|
videos/HRA_part1/2941.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A compassionate woman in a khaki safari outfit gently caresses the muzzle of a rhinoceros, standing on wooden docks overlooking a tranquil lake. The woman, wearing a wide-brimmed hat and a satisfied smile, connects with the rhino, which has a textured hide and a horn. Both are framed by lush greenery and set against a backdrop of serene water, creating a harmonious scene of human-animal interaction. The woman's affectionate touch and the rhino's calm demeanor, along with the peaceful lake in the background, evoke a sense of mutual respect and tranquility.]"
},
{
"from": "gpt",
"value": "Normal. The video clearly shows the woman petting the rhinoceros, the actions are basically normal but there is some lack of coordination."
}
] |
seed42_1638
|
videos/HRA_part0/1638.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a waiter, wearing a black vest and bow tie, is seen playfully crawling on his hands and knees across a wooden floor, following a small, adorable hamster. The hamster, enclosed in a clear plastic ball, rolls ahead of him, creating a comical and heartwarming interaction between the two. The setting is a tidy room with a hardwood floor, adding to the charm of this unexpected pairing.]"
},
{
"from": "gpt",
"value": "Bad. The hamster's face, the person's hand and leg in the video have generation issues, and one of the person's hands suddenly disappears in the middle frame."
}
] |
seed46_2078
|
videos/HRA_part2/2078.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A police officer, clad in a dark uniform with a yellow reflective vest, walks alongside a panther in an urban environment. They navigate through art galleries and exhibition spaces, where abstract paintings and sculptures decorate the walls. The officer's authoritative presence contrasts with the panther's majestic and wild nature, creating an unexpected duo as they stroll through the cultural landscape, surrounded by works of art that range from geometric patterns to lifelike portraits.]"
},
{
"from": "gpt",
"value": "Bad. The head of the black panther, the face of the police officer, and the hands and feet were generated abnormally."
}
] |
seed46_3160
|
videos/HRA_part2/3160.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal scene, a well-lit, multi-level highway with heavy traffic serves as the backdrop. Amidst the urban landscape, a unique character takes center stage—a person dressed in a full-body, realistic cat costume, complete with whiskers and a tail. This feline figure energetically performs a captivating dance routine, drawing attention away from the busy city environment. Despite the unusual setting, the dancer's movements are fluid and expressive, adding a touch of whimsy to the concrete jungle. The contrast between the gritty, urban backdrop and the playful, dancing cat costume creates a memorable visual narrative.]"
},
{
"from": "gpt",
"value": "Bad. The movements are not continuous, there are sudden changes."
}
] |
seed44_1474
|
videos/HRA_part1/1474.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a lush, vibrant forest setting, a mother beaver tenderly caresses her kit with her large, front paw, both sitting cozily within their wooden lodge. The mother, with her brown fur and attentive gaze, gently strokes the smaller beaver, who looks up at her with wide eyes, creating a heartwarming scene of maternal love. The wooden lodge, intricately built from branches and mud, provides a safe and nurturing environment for the family, with green leaves and trees visible in the background, highlighting the beauty of their natural habitat.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed44_2254
|
videos/HRA_part1/2254.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A janitor, middle-aged with short, neatly trimmed brown hair and a well-groomed beard, pauses in his cleaning duties. Dressed in a dark green uniform with a name badge, he rests a mop against a pristine white wall. His eyes meet the camera, and he gives a slow, deliberate nod, acknowledging an unseen presence. The setting is a spotless, modern school hallway, with large windows casting soft light onto the polished floor, highlighting the cleanliness maintained by the janitor.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed44_1118
|
videos/HRA_part1/1118.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged writer, donning a brown jacket, beige slacks, and brown boots, takes a break from his typewriter to rake a large pile of colorful autumn leaves in his backyard, capturing the essence of seasonal change. He pauses, tool in hand, casting a thoughtful gaze at the vintage typewriter sitting on a wooden desk, which is placed under a large, leafy tree. The scene transitions to show him raking leaves with vigor, his attire unchanged, as the leaves form a vibrant heap, evoking a sense of tranquility and the beauty of fall. Finally, he pauses again, this time with a satisfied smile, surveying his neatly raked yard, the pile of leaves a testament to his labor, while the typewriter remains in the background, hinting at his dual passions for writing and nature.]"
},
{
"from": "gpt",
"value": "Bad. The character's movements are a bit discontinuous, the limbs are somewhat uncoordinated, and the hoe in hand undergoes a sudden change."
}
] |
seed42_2931
|
videos/HRA_part0/2931.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A construction site bustling with activity, filled with hard hats and heavy machinery, serves as the backdrop for a massive crane towering overhead. The crane, painted in bold yellow and black stripes, extends its long arm skyward, with a hook dangling below. Sunlight glints off the metal, highlighting the intricate framework and gears that enable the crane's powerful movements. At the top, a small cabin shelters the operator, who skillfully maneuvers the crane to lift and place heavy steel beams, essential for building the city's skyline. Dust and debris swirl around the base, where workers guide the crane's actions with precise hand signals, contributing to the urban landscape's ongoing transformation.]"
},
{
"from": "gpt",
"value": "Good. Basically no problems."
}
] |
seed44_299
|
videos/HRA_part1/299.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A rugged man, resembling a timber wolf with his scruffy beard and wild eyes, leaps through the air in a dramatic display of strength and agility. Dressed in a blue plaid shirt, black pants, and brown work boots, he soars over a pile of logs in a forest clearing. The scene is reminiscent of a werewolf transformation, with the man's intense gaze and powerful muscles highlighting his primal nature. The action takes place under a full moon, casting an eerie glow over the landscape and emphasizing the supernatural theme.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed44_2113
|
videos/HRA_part1/2113.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A janitor, clad in a gray jacket over a white shirt and black pants, is hard at work unclogging a drain in a dimly lit room. He kneels on the floor, his focus intense as he uses a metal tool to clear the clog. The scene is framed by pipes and machinery, emphasizing the industrial setting. His work is meticulous, as he carefully removes debris from the drain, revealing the inner workings of the pipe. Throughout, his dedication to the task is evident, as he ensures the drain is thoroughly cleared, the water now flowing freely, with the satisfaction of a job well done apparent in his demeanor.]"
},
{
"from": "gpt",
"value": "Bad. The actions are not continuous, there are sudden changes."
}
] |
seed44_3009
|
videos/HRA_part1/3009.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic setting, a police officer, clad in a high-tech suit with neon accents, runs across a rugged asteroid surface illuminated by the distant stars. The officer, equipped with a helmet and breathing apparatus, sprints past mining equipment and industrial structures, indicating an asteroid mining operation. With each stride, the officer's suit glows under the harsh lighting of the asteroid's environment, highlighting the action and determination in this dynamic scene set against the backdrop of space exploration and law enforcement.]"
},
{
"from": "gpt",
"value": "Bad. The officer's hand in the video is distorted and deformed."
}
] |
seed44_1001
|
videos/HRA_part1/1001.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A smiling firefighter, clad in full gear and a helmet, stands against the backdrop of a fire station. His turnout coat is unzipped, revealing a yellow safety vest, and he holds his helmet in one hand. The fire engine bay is visible behind him, with a red fire truck and a large American flag hanging on the wall. His cheerful demeanor is accentuated by the bright sunlight shining through the open garage door, casting a warm glow over the scene. Despite the serious nature of his profession, his smile conveys a sense of pride, resilience, and community spirit.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed42_2722
|
videos/HRA_part0/2722.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a lush, green park under the bright summer sun, two happy children, a boy and a girl, are energetically chasing each other on the grass. The boy, wearing a red t-shirt and blue shorts, stretches out his hands, playfully trying to catch the girl. She, donned in a yellow dress and sandals, giggles as she skillfully dodges him, both enjoying the thrill of the chase. They run amidst a scattering of colorful wildflowers, their laughter echoing through the serene environment, encapsulating the joy and innocence of childhood.]"
},
{
"from": "gpt",
"value": "Bad. The video clearly shows the children chasing each other, but the movements are very discontinuous and jerky."
}
] |
seed44_296
|
videos/HRA_part1/296.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A lawyer, dressed in a dark suit, white shirt, and blue tie, finds himself inexplicably transported to a different era, reminiscent of the TV show 'Quantum Leap.' Crawling through a dimly lit, gritty tunnel, he struggles to understand his time-bending predicament. The tunnel walls are adorned with graffiti and debris, adding to the disorienting atmosphere. Suddenly, the lawyer emerges into a bustling courtroom from the past, where he must now navigate unfamiliar legal challenges. His bewildered expression and the stark contrast between his modern attire and the vintage courtroom setting underscore the surreal nature of his quantum leap experience.]"
},
{
"from": "gpt",
"value": "Bad. The character's face, hands, and arms are distorted in the video."
}
] |
seed42_1773
|
videos/HRA_part0/1773.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist, donning a yellow hazmat suit typically worn in industrial areas, paints a vivid mural featuring a graceful gazelle sprinting across a barren urban landscape. The scene is set against the backdrop of industrial zones, with the artist's focused expression and steady hand capturing the gazelle's dynamic movement on a large concrete wall. The artwork comes to life as the painter adds final touches, highlighting the gazelle's speed and agility, juxtaposed against the gritty industrial setting, while the artist, still in the yellow protective gear, steps back to admire the finished piece.]"
},
{
"from": "gpt",
"value": "Normal. The character's hands are distorted in the video."
}
] |
seed44_15
|
videos/HRA_part1/15.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal scene, a young, bespectacled Asian male cashier, wearing a red apron over a gray shirt, stands in a dimly lit grocery store. Behind him hangs a large, vintage television playing footage of the historic moon landing. Suddenly, the focus changes to reveal a majestic deer walking calmly down the aisle, seemingly unfazed by its unusual surroundings. The cashier looks on in mild surprise, his expression a mix of bemusement and acceptance, as the deer continues its leisurely stroll among the shelves.]"
},
{
"from": "gpt",
"value": "Normal. The deer has two heads."
}
] |
seed44_2072
|
videos/HRA_part1/2072.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man, with short brown hair and a neatly trimmed beard, stands in a modern kitchen with white cabinets and countertops. He is wearing a casual outfit consisting of a gray t-shirt and dark blue jeans. The kitchen is well-lit, and natural light filters in from a nearby window. The man is focused on making coffee, pouring water into a sleek, stainless steel coffee maker. He then measures coffee grounds carefully, ensuring the perfect cup. As the coffee brews, steam rises gently, and the aroma fills the room, creating a warm and inviting atmosphere. The man waits patiently, his expression one of anticipation, as the coffee completes its cycle, ready to be enjoyed.]"
},
{
"from": "gpt",
"value": "Bad. The man's hand and the item in his hand were generated incorrectly in the video."
}
] |
seed46_1977
|
videos/HRA_part2/1977.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A kingfisher bird, vibrant in its electric blue plumage and orange breast, perches on a thin branch overhanging a tranquil pond, its sharp eyes intently scanning the water's surface. The sunlight filters through the leaves above, casting dappled light onto the bird and the water below. Suddenly, the kingfisher dives headfirst into the water, its body slicing through the calm surface with precision. It emerges a moment later, triumphant, a small fish wriggling in its beak. The bird then returns to its branch, ready to enjoy its catch, the water still rippling from its dive and the surrounding foliage swaying gently in the breeze.]"
},
{
"from": "gpt",
"value": "Bad. The scene of the kingfisher flying back to the branch with a small fish in its beak, emerging from the water, was not generated."
}
] |
seed44_2477
|
videos/HRA_part1/2477.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An engineer wearing a high-visibility jacket walks along the edge of a bustling harbor, the sunlight glinting off the water and reflecting onto his safety glasses. He carries a clipboard in one hand, attentively observing the large ships and cranes at work. The harbor is alive with activity, from the loading and unloading of cargo to the movement of tugboats guiding massive vessels. In the distance, the silhouette of a city skyline adds a touch of urbanity to the industrial scene. The engineer stops occasionally to jot down notes, his expertise evident in his confident demeanor and the precision with which he performs his duties.]"
},
{
"from": "gpt",
"value": "Normal. The character's movements in the video are slightly unsmooth."
}
] |
seed42_119
|
videos/HRA_part0/119.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a heartwarming scene, two children, a boy and a girl, are seen gently caressing a juvenile antelope in a lush, green meadow. The boy, wearing a blue shirt and shorts, leans over to pet the animal with a tender smile, while the girl, dressed in a yellow dress, watches adoringly, her hands clasped together. Moments later, the girl takes her turn, stroking the antelope's back as it stands calmly, accepting the affection. The boy looks on with a joyful expression, his right hand raised in excitement. The idyllic setting, complete with verdant grass and blue skies, adds to the peaceful and endearing nature of the interaction between the children and the young antelope.]"
},
{
"from": "gpt",
"value": "Good. Basically no problem."
}
] |
seed46_1608
|
videos/HRA_part2/1608.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A mature, bespectacled teacher, wearing a gray blazer over a button-down shirt, walks down a dimly lit, empty hallway with lockers on either side. Suddenly, a small brown rat scurries across the floor in front of him, causing the teacher to pause and look down with a mix of surprise and curiosity. The rodent continues its journey undeterred, disappearing around a corner as the teacher resumes his walk, now with a slight smile and a shake of his head, as if amused by the unexpected encounter.]"
},
{
"from": "gpt",
"value": "Bad. The character in the video walks very unsmoothly, with uncoordinated movements; the mouse's actions are also uncoordinated."
}
] |
seed42_1637
|
videos/HRA_part0/1637.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A golden retriever, framed against a vibrant green meadow under a clear blue sky, experiences a unique perception of time as it runs. Initially, the dog is seen standing still, looking forward with anticipation. It then starts running, its legs blurring as if time has accelerated for the viewer. Next, the dog leaps into the air, seemingly frozen in mid-air, giving the impression of time slowing down. Finally, the dog bounds across the field, its motion smooth and unhurried, conveying a sense of normal time flow, all while the lush green landscape and clear blue sky provide a serene backdrop.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed42_2496
|
videos/HRA_part0/2496.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A sleek, green-brown jungle cobra crawls across a lush, moss-covered log in the dense rainforest undergrowth. The snake's hood is slightly raised, revealing its distinctive markings as it navigates the textured surface. Sunlight filters through the canopy, casting dappled light on the cobra, emphasizing its muscular body and smooth scales as it moves with purpose and grace in its natural habitat.]"
},
{
"from": "gpt",
"value": "Good. No obvious issues."
}
] |
seed44_2952
|
videos/HRA_part1/2952.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished film director, wearing a brown leather jacket and dark pants, kneels on a wooden floor, engaged in a crawling motion, as he observes a live hare sitting on a white table nearby. The scene is set indoors, with a large window in the background, allowing natural light to illuminate the room. The director's intent gaze and careful movement convey a sense of focus and determination as he studies the animal, which remains calm and still, seemingly unfazed by the presence of the human.]"
},
{
"from": "gpt",
"value": "Normal. The director's face is distorted in the video."
}
] |
seed46_1890
|
videos/HRA_part2/1890.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A janitor, clad in a green vest and khaki pants, walks purposefully down a dimly lit hallway, pushing a squeaky cleaning cart. On the wall, a badger mascot poster from a local college, Wausau Woodchucks, adds a touch of whimsy. The scene shifts slightly to reveal more of the hallway, showcasing the linoleum floor and overhead lights. The janitor continues his journey, passing a row of lockers adorned with stickers and graffiti, while the badger poster remains visible, adding a sense of character to the otherwise mundane setting. Finally, the focus is on the janitor's interaction with the environment, as he pushes the cart with determination, the badger poster serving as a silent observer in the background.]"
},
{
"from": "gpt",
"value": "Normal. The face, hands, and legs of the cleaner in the video are generated incorrectly."
}
] |
seed44_2986
|
videos/HRA_part1/2986.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A young, cheerful nurse wearing a green uniform jumps playfully over a fallen log in the lush rainforest, surrounded by vibrant green foliage, exotic flowers, and towering trees. She lands gracefully on the moss-covered ground, her smile conveying a sense of freedom and joy. Nearby, a small waterfall cascades into a crystal-clear pool, adding to the serene atmosphere of the tropical environment. The nurse's unexpected presence in the rainforest, combined with her playful leap, creates a captivating and memorable scene.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed42_1385
|
videos/HRA_part0/1385.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A close-up view of a whole, raw Atlantic cod fish, eyes glassy and intact, laid upon a bed of ice chips, its skin shimmering under bright market lights. The fish is sizable, with a length suggesting it weighs several pounds, and is presented head-on, showcasing its freshness in a fishmonger's display. The setting is a bustling, open-air market, with other seafood visible in the background, emphasizing the variety and vibrancy of the scene. The cod's mouth is slightly agape, and its body lies across the ice, evoking the traditional presentation of fresh catches in coastal communities.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed44_3121
|
videos/HRA_part1/3121.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dense forest setting, a middle-aged man dressed in a green jacket and gray pants is seen wiping a wooden signpost with a cloth. The signpost, which bears the name of the forest trail, is set against a backdrop of lush greenery and tall trees. The man, likely a park ranger or caretaker, focuses intently on his task, ensuring the sign is clean and legible for visitors. The scene captures a moment of maintenance in the midst of nature's beauty.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed42_2265
|
videos/HRA_part0/2265.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a bustling mall filled with shoppers, a courageous firefighter clad in full gear, including a helmet and breathing apparatus, runs purposefully down a wide corridor. The vibrant red of his turnout coat contrasts sharply with the gray floor tiles and the bright storefronts. Escalators and elevators stand as silent witnesses to his urgent mission, while the clean, contemporary lines of the mall's architecture provide a stark backdrop to the unfolding drama. Amidst the commercial ambiance, the firefighter's presence and haste underscore the gravity of the situation, his figure growing smaller as he disappears into the distance, leaving behind a trail of determination and heroism.]"
},
{
"from": "gpt",
"value": "Bad. The character's movements in the video are not smooth."
}
] |
seed44_1552
|
videos/HRA_part1/1552.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen note that due to the rule about scene transitions, the last sentence should be disregarded, so here's an alternative ending without a transition:\nThe camera then pans out to show the sand dollar resting peacefully on the sandy seabed, surrounded by a few small pebbles and seaweed, capturing the serenity of its underwater habitat. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A close-up view captures the intricate beauty of a sand dollar, its pale color and fine texture contrasting against a dark, grainy sand background. The camera zooms in to reveal the small, delicate spines covering the surface of the sand dollar, emphasizing its natural design. As the focus shifts, the circular shape of the sand dollar becomes prominent, highlighting its symmetrical pattern. The final shot showcases the sand dollar partially buried in the sand, with a blurred oceanic horizon in the distance, evoking a sense of mystery and wonder about this enigmatic sea creature.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed46_1114
|
videos/HRA_part2/1114.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a captivating underwater scene, a cluster of Moon jellyfish gracefully float in a horizontal orientation, their translucent bell-shaped bodies appearing white with a hint of blue. The jellyfish showcase their four reproductive structures, known as gonads, which resemble red or purple petals. Surrounding these central features are delicate, trailing tentacles in a darker shade. The setting is enhanced by the soft glow of ambient light filtering through the water, creating an ethereal atmosphere as the jellyfish slowly undulate in the current, evoking a sense of peaceful wonder.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed44_1932
|
videos/HRA_part1/1932.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An inventive engineer, wearing safety goggles and a lab coat, has designed a unique, waterproof hamster-sized diving suit. The hamster, now transformed into an aquatic explorer, swims energetically in a clear fish tank, bubbles rising around it. The tank is placed on a wooden table in a cluttered workshop filled with tools and gadgets. The engineer watches intently, clipboard in hand, documenting the hamster's swimming patterns as it propels itself through the water with its tiny limbs, creating a whimsical yet scientific scene.]"
},
{
"from": "gpt",
"value": "Bad. No generation of engineers wearing safety goggles and lab coats, waterproof diving suits for hamsters, or writing boards."
}
] |
seed46_2095
|
videos/HRA_part2/2095.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit room filled with the scent of old books and memories, two people sit on a vintage, worn-out sofa, their faces reflecting deep sorrow. One, a woman with long, wavy brown hair, wears a simple gray sweater and blue jeans, her eyes red from crying, as she holds a well-loved teddy bear. Beside her, a man in a plaid shirt and khaki pants, his face equally stricken with grief, leans his forehead against hers, both of them finding solace in each other's presence. Tears stream down their cheeks, highlighting the raw emotion of the moment, as they seek comfort in the quiet space they share.]"
},
{
"from": "gpt",
"value": "Normal. The woman's hand appears deformed and distorted."
}
] |
seed42_182
|
videos/HRA_part0/182.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vibrant community garden, a group of diverse individuals, including a woman with brown hair tied in a bun, wearing a gray jacket and blue jeans, and a bald man in a red T-shirt and blue jeans, work together to pull weeds from a raised flower bed. The scene is set against a backdrop of lush greenery and blooming flowers, with a wooden shed and a painted mural visible in the distance. The cooperative atmosphere is highlighted by the shared effort of the group, each member engaged in the task of maintaining the garden's beauty.]"
},
{
"from": "gpt",
"value": "Bad. People's hands appear deformed, the woman's arm is distorted, and the man has a third hand. People in the background also show deformities, some without heads."
}
] |
seed44_1802
|
videos/HRA_part1/1802.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A college student, wearing a red and white striped shirt, dark pants, and a vibrant tiger head costume, energetically dances in front of a large crowd at a lively pep rally. The atmosphere is electric, filled with cheering fans and the spirit of school pride. The young man, donning the mascot's full-body suit complete with furry orange and black stripes, performs dynamic dance moves to the beat of an upbeat marching band, captivating the audience with his animated antics. Surrounded by enthusiastic supporters waving banners and flags, the scene pulses with excitement and unity, perfectly capturing the essence of collegiate spirit.]"
},
{
"from": "gpt",
"value": "Bad. The character's movements in the video are not smooth."
}
] |
seed42_1587
|
videos/HRA_part0/1587.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A vivid close-up of a grouse's head showcases its distinctive red facial wattles and curious, alert expression, set against a blurred forest background. The bird's intricate feather patterns are highlighted, particularly the dark speckles on its brownish-grey plumage. Its sharp, intelligent eyes and slightly open beak convey a sense of awareness and readiness in its woodland habitat. The grouse is perched on a branch, surrounded by lush greenery, with rays of sunlight filtering through the leaves, illuminating its feathers and the rugged bark of the tree.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed42_2418
|
videos/HRA_part0/2418.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit room, a focused doctor, dressed in a white lab coat, leans over a table, examining a specimen. On the floor below, a tiny mouse scurries across the wooden surface, navigating the shadows beneath the furniture. The scene alternates between the doctor's intense study and the mouse's cautious movements, highlighting the stark contrast between the clinical environment and the unexpected presence of the small creature.]"
},
{
"from": "gpt",
"value": "Bad. The mouse in the video walks very unsmoothly, with uncoordinated movements."
}
] |
seed42_2078
|
videos/HRA_part0/2078.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A police officer, clad in a dark uniform with a yellow reflective vest, walks alongside a panther in an urban environment. They navigate through art galleries and exhibition spaces, where abstract paintings and sculptures decorate the walls. The officer's authoritative presence contrasts with the panther's majestic and wild nature, creating an unexpected duo as they stroll through the cultural landscape, surrounded by works of art that range from geometric patterns to lifelike portraits.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed46_2427
|
videos/HRA_part2/2427.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist dons a spacesuit and helmet, adorned with vibrant colors and patterns, resembling a solar sail, as they crawl across a barren lunar landscape, backlit by the Earth's glow. The suit, with its reflective, sail-like design, catches the sunlight, creating mesmerizing reflections and shadows. The artist moves slowly, emphasizing the suit's unique texture and the dramatic contrast between the suit and the moon's surface. This artistic portrayal on the moon's surface, with the distant Earth providing a serene backdrop, highlights the intersection of art and space exploration.]"
},
{
"from": "gpt",
"value": "Normal. The artist's limbs in the video have generation issues, with the body being deformed."
}
] |
seed42_18
|
videos/HRA_part0/18.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A North American porcupine, identified by its dark brown fur and quills, is seen jumping with surprising agility from branch to branch in a dense forest. The scene captures the porcupine's unique climbing skills as it leaps from one limb to another, highlighting its ability to navigate its arboreal habitat. The video focuses on the porcupine's movements, emphasizing the contrast between its slow, deliberate gait when on solid ground and its sudden bursts of energy as it propels itself through the air. The forest surroundings provide a natural setting, with sunlight filtering through the canopy to illuminate the porcupine's quills during its jumps.]"
},
{
"from": "gpt",
"value": "Normal. The hedgehog's nose in the video is fully generated, but there are missing parts when the spikes on its body overlap with the branches."
}
] |
seed46_1859
|
videos/HRA_part2/1859.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A uniformed police officer kneels on a grassy park area, attentively watching a squirrel. The squirrel, emboldened, leaps from the ground onto the officer's outstretched arm, seeking a nut held in the officer's other hand. The cop wears a standard police uniform, including a navy blue shirt, black tactical pants, and a badge, while the greenery and trees in the background set a peaceful outdoor scene. The squirrel's acrobatic leap and the officer's gentle interaction create a heartwarming moment in this unexpected encounter.]"
},
{
"from": "gpt",
"value": "Bad. The scene of the squirrel jumping onto the policeman's outstretched arm was not generated;"
}
] |
seed42_1767
|
videos/HRA_part0/1767.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A solitary buzzard perches majestically on a barren branch, its piercing eyes scanning the vast, sunlit savannah below for any signs of movement. The bird, with its distinct brown and white plumage, contrasts starkly against the muted tones of the landscape. Suddenly, it spreads its impressive wingspan and takes flight, gliding effortlessly over the terrain. Its sharp talons are clearly visible as it descends with laser-like precision towards a potential meal, showcasing the raw power and agility of this apex predator. The scene is set against a backdrop of rolling hills and scattered trees, bathed in the warm glow of the afternoon sun.]"
},
{
"from": "gpt",
"value": "Bad. The character's movements in the video are unsmooth."
}
] |
seed46_302
|
videos/HRA_part2/302.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged man, donning a yellow apron over his gray shirt and dark pants, diligently scrubs a bathroom floor with a green cleaning solution, using a large, round scrub brush. The bathroom is modern, featuring white walls, a white toilet, and a white sink. The man's focused expression and the steam from the hot water convey his determination to clean. In the next shot, he is seen squeezing out the dirty water from the mop into a large, rectangular bucket. Finally, he pauses to wipe his forehead with a red cloth, revealing his dedication to maintaining cleanliness in the restroom.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed44_1339
|
videos/HRA_part1/1339.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man with a muscular build, wearing a white sleeveless undershirt and dark gray jeans, stands on a wooden platform, intently applying paint to a large canvas with a brush. The focus is on his hands, which are covered in paint, as they skillfully create an abstract artwork. Behind him, a row of neatly arranged paint cans sits on a wooden floor, and the walls of the studio are painted white, creating a clean, professional environment. The artist's concentration is evident in his furrowed brow and the intensity of his gaze, while the vibrant colors on the canvas contrast with the neutral tones of his clothing and surroundings.]"
},
{
"from": "gpt",
"value": "Good. Basically no problem."
}
] |
seed46_1588
|
videos/HRA_part2/1588.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vast, sunlit savannah, two zebras engage in a tender moment, their striped bodies gently touching as they caress each other. One zebra, with its mane flowing in the breeze, nuzzles the face of another, their bond evident in the softness of their interaction. The scene is framed by distant acacia trees and a clear blue sky, emphasizing the intimacy shared between these majestic creatures in their natural habitat.]"
},
{
"from": "gpt",
"value": "Normal. Zebra's head is distorted, and its legs are twisted."
}
] |
seed44_1119
|
videos/HRA_part1/1119.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A cheerful Asian cashier, wearing a red apron over a white collared shirt, black pants, and black shoes, is seen mid-air, jumping with both feet off the ground, her ponytail bouncing behind her. The camera captures her from various angles, showing her enthusiasm and energy as she performs this unexpected action within the confines of her workplace, a modern convenience store with neatly arranged shelves and bright lighting.]"
},
{
"from": "gpt",
"value": "Bad. The woman's limbs in the video are generated abnormally, her face is blurry, and her body is distorted."
}
] |
seed46_1288
|
videos/HRA_part2/1288.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A dingo, exhibiting a reddish-brown coat and alert demeanor, crawls near the base of a lush, cascading waterfall set within a dense, verdant forest. The waterfall, framed by towering trees and vibrant foliage, provides a dramatic backdrop as it plunges into a clear pool below, where the dingo stealthily approaches. The animal's cautious movement, combined with the serene yet powerful presence of the waterfall, creates a captivating scene of nature's raw beauty and untamed spirit.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed42_2714
|
videos/HRA_part0/2714.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic classroom aboard a space station, a compassionate teacher, dressed in a blue jumpsuit with patches and pins adorning her chest, gently caresses the head of a curious alien student. The student, with large, expressive eyes and a small, round body, sits attentively at a desk cluttered with advanced technology and educational tools. The room is filled with natural light from a large window showing a stunning view of Earth and the vast cosmos beyond, emphasizing the peaceful diplomatic mission of education among the stars.]"
},
{
"from": "gpt",
"value": "Normal. The video clearly shows the teacher touching the student's head; the action is basically normal but slightly uncoordinated."
}
] |
seed42_2026
|
videos/HRA_part0/2026.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A magnificent cheetah, the world's fastest land animal, is shown in a lush, green savanna environment. The big cat is engaged in a self-cleaning ritual, using its paw to wipe across its face in a natural grooming behavior. The cheetah's iconic spotted coat is visible as it sits on the grassy ground, performing this common yet intimate act. The scene captures the cheetah's solitary nature and the importance of hygiene in its daily life, all set against a backdrop of vibrant vegetation and clear skies.]"
},
{
"from": "gpt",
"value": "Good. No obvious issues."
}
] |
seed46_2603
|
videos/HRA_part2/2603.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit, smoke-filled environment, a firefighter wearing full gear, including a helmet and breathing apparatus, crawls through a confined space. The setting is a high-tech facility, indicated by the presence of an antimatter containment unit, which is a large, cylindrical container with intricate machinery and glowing blue lights, seen in the background. The firefighter's movements are slow and deliberate, emphasizing the hazardous conditions and the need for caution. Despite the challenging circumstances, the firefighter's determination is evident, as they navigate the narrow passage with a flashlight in hand, illuminating the path ahead. The overall scene conveys a sense of urgency and danger, with the firefighter's mission being critical to the safety of the facility and its personnel.]"
},
{
"from": "gpt",
"value": "Normal. The character's movements in the video are slightly unsmooth."
}
] |
seed42_1501
|
videos/HRA_part0/1501.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A police officer, clad in a navy uniform with a badge and a tan belt holding a radio and handcuffs, diligently repairs a damaged wooden fence in a suburban neighborhood. The fence, marred by graffiti and broken planks, is set against a clear blue sky. The officer, wearing a hat and sunglasses, demonstrates community involvement by hammering a new plank into place, emphasizing public service beyond traditional law enforcement duties.]"
},
{
"from": "gpt",
"value": "Normal. The officer's movements in the video are basically normal, with slight awkwardness."
}
] |
seed44_2926
|
videos/HRA_part1/2926.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A well-dressed lawyer, wearing a gray suit, white shirt, and striped tie, stands on a chair in a modern office space with high ceilings and large windows. Holding a light bulb, he appears mid-action, changing the bulb with determination. The office is filled with contemporary furniture and technology, including a sleek desk, ergonomic chair, and flat-screen monitor. The scene is lit by the natural light streaming in from the windows, casting a warm glow on the lawyer as he completes his unexpected task.]"
},
{
"from": "gpt",
"value": "Normal. The lawyer's movements in the video are uncoordinated."
}
] |
seed44_2903
|
videos/HRA_part1/2903.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A young female cashier, wearing a red apron over a white collared shirt, black pants, and brown shoes, pulls a heavy, overflowing shopping cart across a busy supermarket floor. She moves with determination, navigating the aisles filled with various products and customers. The cart, brimming with goods, contrasts with her clean and tidy appearance, highlighting the physical demands of her job. Behind her, the supermarket's brightly lit shelves and bustling atmosphere provide a dynamic backdrop to her task.]"
},
{
"from": "gpt",
"value": "Good. No obvious problems."
}
] |
seed46_2112
|
videos/HRA_part2/2112.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a softly lit room, a gentle hand caresses a small, curious mouse, creating a tender scene. The mouse, with its tiny body and delicate features, sits on a smooth wooden surface, possibly a desk or table. The hand, belonging to a caring individual, moves gently, stroking the mouse's back with slow, soothing motions. The atmosphere is peaceful, with the focus solely on the interaction between the human and the mouse, emphasizing the bond formed in this quiet moment. The mouse appears relaxed and comfortable under the hand's touch, highlighting the trust and connection that has developed between them.]"
},
{
"from": "gpt",
"value": "Good. No obvious issues."
}
] |
seed42_2961
|
videos/HRA_part0/2961.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A construction worker, clad in a yellow vest and hard hat, finds a moment of joy as he dances alone in the vast, sun-baked expanse of a subtropical desert. The dunes stretch endlessly under a cloudless sky, the horizon melding seamlessly into the distance. His movements, expressive and free, contrast sharply with the stillness of the arid landscape. The golden sands ripple beneath his boots, each step a fleeting imprint in the relentless expanse. As he twirls and leaps, the desert becomes his stage, the unyielding sun his spotlight, capturing a fleeting moment of uninhibited happiness in an otherwise harsh environment.]"
},
{
"from": "gpt",
"value": "Bad. The man in the video is generally distorted."
}
] |
seed44_2529
|
videos/HRA_part1/2529.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A young female cashier with brown hair pulled into a bun, wearing a red apron over a white collared shirt, serves customers at a grocery store checkout lane. Suddenly, her eyes well up with tears, and she begins to quietly cry, her face showing a mix of sadness and professionalism as she continues ringing up items. The scene highlights her emotional struggle, balancing work responsibilities with personal feelings, all under the fluorescent lights of the supermarket.]"
},
{
"from": "gpt",
"value": "Bad. The woman's eyes in the video were not generated."
}
] |
seed46_2016
|
videos/HRA_part2/2016.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal scene, a majestic gorilla, adorned with a vibrant, golden crown and a silver cape, dances gracefully on a floating platform set within Neptune's ethereal atmosphere. The planet's vivid blue and green gaseous clouds swirl dramatically in the background, creating an otherworldly backdrop. The gorilla performs a series of elegant dance moves, its powerful yet gentle demeanor captivating the viewer. The platform, illuminated by a soft, celestial glow, adds to the mystical ambiance of this extraordinary interstellar dance performance.]"
},
{
"from": "gpt",
"value": "Bad. The gorilla's movements in the video are very unsmooth and uncoordinated."
}
] |
seed44_2176
|
videos/HRA_part1/2176.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a captivating display, a slender cheetah with golden fur and black spots leaps into the air against a vivid, sunlit savanna landscape. Its body curves gracefully mid-leap, capturing the essence of its feline agility. The cheetah's intense, focused gaze is directed forward, suggesting a sense of purpose in its dance-like movements. Its tail extends elegantly, balancing its airborne form as it seemingly dances across the plains, embodying the spirit of the wild. The scene is framed by distant acacia silhouettes, enhancing the dramatic effect of the cheetah's dynamic leap.]"
},
{
"from": "gpt",
"value": "Good. Basically no issues."
}
] |
seed44_2828
|
videos/HRA_part1/2828.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A uniformed police officer, donning a navy blue cap and sunglasses, runs along the sandy shore of a bustling beach resort, where vibrant umbrellas and lively beachgoers fill the background. His attire includes a navy blue shirt with badge, navy blue trousers, and black shoes. The setting is a crowded beach during daytime, with the officer's movement capturing the essence of duty amidst leisure. He sprints past colorful umbrellas and animated vacationers, highlighting the contrast between his purposeful stride and the relaxed atmosphere of the resort.]"
},
{
"from": "gpt",
"value": "Bad. The video clearly shows the police running on the beach, but the movements are not very continuous and there is a lack of coordination."
}
] |
seed44_2658
|
videos/HRA_part1/2658.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished film director, donning a dark green shirt, beige shorts, and a brown belt, is seen meticulously dusting vintage furniture within a grand room filled with antique artifacts. He pauses to adjust his glasses, revealing a look of deep concentration. His gaze then shifts to a large painting hanging on the wall, as if contemplating its significance. The director continues his task, now focusing on a wooden cabinet, his movements deliberate and precise, demonstrating his attention to detail. Lastly, he inspects a polished table, his reflection visible in its gleaming surface, as he ensures every inch is free from dust, all the while remaining in the same room with its collection of cherished items.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed44_1774
|
videos/HRA_part1/1774.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surprising scene set within the lush environment of a tropical dry forest, a young, smiling female cashier, wearing a red apron over a white blouse and black pants, energetically jumps over a small log. She lands gracefully, holding a miniature baseball bat, as if ready for an unexpected game amidst the dense vegetation. The vibrant green foliage and dappled sunlight create a vivid contrast against her outfit, capturing the spirit of adventure and playfulness in an unconventional setting.]"
},
{
"from": "gpt",
"value": "Bad. The character in the video has a twisted body, face, arms, legs, and feet."
}
] |
seed44_2561
|
videos/HRA_part1/2561.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A film director, clad in a dark green jacket, blue jeans, and a beige hat, walks through a dry, rocky landscape, his attention fixed on a hyena in the distance. The hyena, displaying its characteristic spotted coat and large ears, roams freely in the arid environment. The director, holding a megaphone and script, seems to be guiding the action or observing the hyena's natural behavior, capturing the essence of the wild animal in a cinematic exploration. The scene conveys a sense of adventure and creativity as the director integrates the captivating presence of the hyena into his storytelling.]"
},
{
"from": "gpt",
"value": "Good. Basically correct."
}
] |
seed42_3054
|
videos/HRA_part0/3054.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist, donning a black beret, glasses, and a striped shirt, stands in a dimly lit studio with paintings in the background. He is engaged in an unexpected task, unclogging a drain, using a plunger on a white ceramic sink, with the focus on his expressive face and the effort exerted in this mundane activity. The contrast between his creative attire and the ordinary plumbing task creates a compelling narrative. Moments later, he pauses, revealing a look of determination as he continues his efforts to clear the clogged drain, the camera capturing his every detail in high definition.]"
},
{
"from": "gpt",
"value": "Normal. The character's movements are not smooth."
}
] |
seed46_2614
|
videos/HRA_part2/2614.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man with a muscular build, wearing a black tank top and blue jeans, pushes against a large, gray boulder in the middle of a lush forest. His face contorts with effort, highlighting his determination. The scene is set in a dense woodland area, with sunlight filtering through the trees and casting dappled shadows on the ground. The man's strong physique and focused expression convey a sense of strength and perseverance as he exerts himself against the immovable object.]"
},
{
"from": "gpt",
"value": "Bad. The man's fingers in the video are missing, the face is blurry, and there is a model penetration with the stone, and the arm generation is unreasonable."
}
] |
seed42_1817
|
videos/HRA_part0/1817.mp4
|
[
{
"from": "human",
"value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A majestic falcon soars through a cloud-filled sky, its piercing eyes and hooked beak sharply defined against the light. It dives gracefully, wings tucked, chasing an unseen prey over lush green fields dotted with wildflowers. The bird, adorned with a leather hood and jesses, lands on a gloved hand, presenting its fierce gaze and powerful talons. Its plumage, a mix of brown, white, and black, contrasts beautifully with the serene landscape, creating a stunning portrait of nature's hunter against a backdrop of rolling hills and a clear blue horizon.]"
},
{
"from": "gpt",
"value": "Bad. The falcon's wings are twisted, the claws are twisted, and the face is slightly distorted."
}
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.