Dataset Viewer
Auto-converted to Parquet
Path
stringlengths
22
36
Category
stringclasses
5 values
Label
stringlengths
7
223
Noise Density
int64
10
90
Speckle Size
int64
1
3
Frame Count
int64
37
1.86k
FPS
float64
17.8
60
Duration
float64
1
35
Movement Type
stringclasses
2 values
video
video
words/0ba9f2b37ab3.mp4
words
['quantum computers calculate']
50
3
131
39.23
3.34
static
words/634238bc4906.mp4
words
['sunflower fields wave']
50
3
118
36.97
3.19
static
words/04f886cbeb6e.mp4
words
['nebulae glow bright']
50
3
149
29.8
5
static
words/69fbc63416fa.mp4
words
['cosmic wind blows']
50
3
150
30
5
static
words/39a316fd07fd.mp4
words
['summer grass sways']
50
3
175
35.21
4.97
static
words/00f1b8d9e4ed.mp4
words
['thunderstorm brewing']
50
3
88
33.05
2.66
static
words/6db3ded0423a.mp4
words
['silver moonbeams']
50
3
167
33.55
4.98
static
words/7c8360ec2466.mp4
words
['aurora lights dance']
50
3
167
33.66
4.96
static
words/dc6e8c60ba9c.mp4
words
['time crystals form']
50
3
173
34.89
4.96
static
words/a7f78cfe2974.mp4
words
['quantum bits flip']
50
3
170
34.24
4.97
static
words/9a83d0fe51ed.mp4
words
['quantum dots align']
50
3
169
33.97
4.98
static
words/ac4f287587c9.mp4
words
['helium']
50
3
157
36.29
4.33
static
words/969d477ec409.mp4
words
['floating paper lanterns']
50
3
169
33.92
4.98
static
words/4d85ae63722e.mp4
words
['gentle stream bubbles']
50
3
176
35.42
4.97
static
words/550975c7d7fe.mp4
words
['evening shadows lengthen']
50
3
163
34.6
4.71
static
words/c2cd39d8ce54.mp4
words
['gentle rain falls']
50
3
146
33.36
4.38
static
words/b1c869d30e0d.mp4
words
['thunder rolls distant']
50
3
40
39.07
1.02
static
words/4458e1b7d3c8.mp4
words
['meteor showers fall']
50
3
184
37.01
4.97
static
words/89e41e7b99f7.mp4
words
['moonlit garden path']
50
3
166
33.35
4.98
static
words/965e65ed6029.mp4
words
['ancient stone walls']
50
3
168
33.79
4.97
static
words/1747a9a2e512.mp4
words
['coffee shop musings']
50
3
204
41.15
4.96
static
words/25ef525f2d38.mp4
words
['silicon dreams process']
50
3
176
35.47
4.96
static
words/e0b6003f160b.mp4
words
['coffee grounds swirl']
50
3
169
34.05
4.96
static
words/5d9d0dab30f7.mp4
words
['cobalt']
50
3
168
33.86
4.96
static
words/133df1827792.mp4
words
['gentle breeze']
50
3
173
34.78
4.97
static
words/3377e2d04bbc.mp4
words
['golden sunset glow']
50
3
180
36.17
4.98
static
words/57ccd8e5370e.mp4
words
['scattered puzzle pieces']
50
3
163
32.77
4.97
static
words/4a75945347ae.mp4
words
['twisted metal dreams']
50
3
107
32.63
3.28
static
words/9bbfb62d515e.mp4
words
['plasma storm rages']
50
3
100
35.6
2.81
static
words/9fff96c18602.mp4
words
['marble halls gleam']
50
3
69
35.51
1.94
static
words/ce13b4d6254f.mp4
words
['wild horses run']
50
3
176
35.52
4.95
static
words/77f831a19865.mp4
words
['artificial minds think']
50
3
184
37.06
4.96
static
words/f6244f4c8ed7.mp4
words
['thunder rolling distant']
50
3
183
36.79
4.97
static
words/a6ac81c738f3.mp4
words
['gravitational waves ripple']
50
3
164
32.95
4.98
static
words/3107f573d1fd.mp4
words
['quantum leap forward']
50
3
165
33.3
4.96
static
words/5dfb3cdeb333.mp4
words
['copper']
50
3
169
34.02
4.97
static
words/29534cf183dd.mp4
words
['dancing shadows beneath stars']
50
3
174
34.94
4.98
static
words/00c2f2a34cca.mp4
words
['velvet darkness falls']
50
3
173
34.86
4.96
static
words/b6ade7832d34.mp4
words
['copper kettle steam']
50
3
165
33.23
4.97
static
words/4a3e6a5b847e.mp4
words
['comet tail streams']
50
3
174
34.99
4.97
static
words/df3e923389b7.mp4
words
['maple leaves turning']
50
3
110
34.43
3.2
static
words/07226fd98a0f.mp4
words
['peaceful lake ripples']
50
3
178
35.82
4.97
static
words/d264e202c6a8.mp4
words
['morning mist rises']
50
3
92
37.19
2.47
static
words/b85d26e24e20.mp4
words
['stellar nurseries birth']
50
3
166
33.46
4.96
static
words/8ac4fd3fba8b.mp4
words
['cosmic dust settles']
50
3
175
35.17
4.98
static
words/d96ea3c9831c.mp4
words
['event horizons form']
50
3
153
33.92
4.51
static
words/261766a8a993.mp4
words
['mountain peaks']
50
3
181
36.43
4.97
static
words/d1e35e69f15d.mp4
words
['neutron stars collide']
50
3
152
33.95
4.48
static
words/e16d0f7223c8.mp4
words
['broken clock ticks']
50
3
157
31.59
4.97
static
words/ed8b7bb85087.mp4
words
['peaceful forest clearing']
50
3
182
36.62
4.97
static
words/97182969b6b2.mp4
words
['starlight bends time']
50
3
57
35.79
1.59
static
words/97b79e05f460.mp4
words
['candlelight flickers softly']
50
3
178
35.92
4.96
static
words/6ab82bd63991.mp4
words
['quantum mechanics work']
50
3
181
36.56
4.95
static
words/23fd9db5eefc.mp4
words
['space debris orbits']
50
3
43
37.16
1.16
static
words/7dcb313cb2fd.mp4
words
['cosmic ice forms']
50
3
171
34.35
4.98
static
words/e11079c52de4.mp4
words
['railway station memories']
50
3
157
31.53
4.98
static
words/9f99f5c93ed5.mp4
words
['melting snow drips']
50
3
180
36.21
4.97
static
words/0c4d460da2dd.mp4
words
['holographic displays flicker']
50
3
168
33.74
4.98
static
words/b9461962f0b8.mp4
words
['piano keys sing']
50
3
167
33.55
4.98
static
words/5a74ac008db0.mp4
words
['desert wind howls']
50
3
166
33.43
4.97
static
words/c85d10baff0d.mp4
words
['quasar signals pulse']
50
3
149
30
4.97
static
words/d080e56f5bdb.mp4
words
['midnight jazz']
50
3
37
37
1
static
words/d2a696482b9d.mp4
words
['neon signs buzz']
50
3
176
35.35
4.98
static
words/74df7e9a6d84.mp4
words
['parallel worlds shift']
50
3
63
40.65
1.55
static
words/b41a5cd33e22.mp4
words
['wind chimes tinkle']
50
3
193
38.83
4.97
static
words/185e3ebbe5f9.mp4
words
['silver moonlight']
50
3
162
32.64
4.96
static
words/9b1e76ae2d6f.mp4
words
['carbon']
50
3
182
36.69
4.96
static
words/f7249a25403d.mp4
words
['crystal clear water']
50
3
163
33.76
4.83
static
words/7358765b3955.mp4
words
['paper planes soar']
50
3
182
36.69
4.96
static
words/defbd74954fb.mp4
words
['cosmic rays penetrate']
50
3
168
33.83
4.97
static
words/69c08188d0ca.mp4
words
['fireflies dance tonight']
50
3
184
37.04
4.97
static
words/2206decbeb88.mp4
words
['autumn leaves falling']
50
3
175
35.26
4.96
static
words/cc00014f593d.mp4
words
['mysterious']
50
3
189
38.06
4.97
static
words/176c76a17514.mp4
words
['crisp apple falling']
50
3
152
38.25
3.97
static
words/59439bc5b940.mp4
words
['mountain cabin smoke']
50
3
108
36.29
2.98
static
words/81554a95393c.mp4
words
['velvet shadow dance']
50
3
169
36.4
4.64
static
words/16f10ef430b4.mp4
words
['mountain stream sings']
50
3
218
43.83
4.97
static
words/7a5b50732360.mp4
words
['mercury']
50
3
174
35.07
4.96
static
words/7b33d1c6f526.mp4
words
['silver']
50
3
152
38.58
3.94
static
words/801fe8afe4c3.mp4
words
['dancing firelight']
50
3
178
35.83
4.97
static
words/e8295d1909f5.mp4
words
['butterfly wings flutter']
50
3
158
35.84
4.41
static
words/4f15ce5d1c1f.mp4
words
['forest creatures gather']
50
3
183
36.74
4.98
static
words/8123396da55d.mp4
words
['copper wires spark']
50
3
174
35.07
4.96
static
words/578d9d20c588.mp4
words
['laser beams cross']
50
3
181
36.35
4.98
static
words/76c08c7163ea.mp4
words
['nova flash brightens']
50
3
184
37.05
4.97
static
words/9d9dd37ecfe8.mp4
words
['dragon scales shimmer']
50
3
43
37.16
1.16
static
words/1b3df5a029af.mp4
words
['electronic pulse']
50
3
46
37.55
1.23
static
words/68df241dda70.mp4
words
['solar winds blow']
50
3
176
35.57
4.95
static
words/cbb558ac2fdc.mp4
words
['neural networks expand']
50
3
169
34.07
4.96
static
words/347eacb2c7c4.mp4
words
['peaceful meditation garden']
50
3
169
34.16
4.95
static
words/f1fb1ef49aa5.mp4
words
['autumn leaves drift']
50
3
168
33.76
4.98
static
words/359d1c6d0008.mp4
words
['winter snow falls']
50
3
194
39.05
4.97
static
words/6d3b071c89c9.mp4
words
['urban jungle pulse']
50
3
181
36.44
4.97
static
words/07143a0a5207.mp4
words
['mountain stream memories']
50
3
167
33.55
4.98
static
words/cebc2f70a0ec.mp4
words
['dimensional gates crack']
50
3
167
33.52
4.98
static
words/f637409b20ec.mp4
words
['pulsar beams flash']
50
3
183
36.85
4.97
static
words/73807ecf93e7.mp4
words
['marble']
50
3
176
35.56
4.95
static
words/dba2b01eb712.mp4
words
['windchimes dance gently']
50
3
167
33.73
4.95
static
words/daccd690eb94.mp4
words
['wildflowers sway gently']
50
3
183
36.75
4.98
static
words/742b0fddcd50.mp4
words
['fresh cut grass']
50
3
182
36.56
4.98
static
End of preview. Expand in Data Studio

SpookyBench-v2

Dataset Description

SpookyBench-v2 is the updated version of spooky-bench benchmark, designed to evaluate vision-language models' capabilities for temporal understanding. The dataset contains both static and dynamic videos with various noise densities and speckle sizes, specifically designed to test models' temporal perception abilities. Check our paper for more details.

Key Features

  • Human accuracy around 98% while all of the models including closed source one gets 0% accuracy.
  • Noise Robustness Testing: Videos with varying noise densities (10-90) and speckle sizes
  • Temporal Understanding: Mix of static and dynamic content to test temporal perception
  • Diverse Categories: Images, words, shapes, real-world videos, and dual-object scenarios
  • Movement Type Classification: Explicit labeling of static vs. dynamic content

Dataset Statistics

  • Total Videos: 1,127
  • Total Frames: 451,649
  • Static Frame Videos: 394
  • Dynamic Videos: 733

Categories Breakdown

  • dual_images: 108 videos
  • images: 404 videos
  • shapes: 28 videos
  • videos: 57 videos
  • words: 530 videos

Dataset Structure

  SpookyBenchDatasets/ 
  β”œβ”€β”€ metadata.csv # Dataset metadata with all annotations 
  β”œβ”€β”€ images/ # Static single-object videos 
  β”œβ”€β”€ words/ # Static word videos 
  β”œβ”€β”€ shapes/ # Static geometric shape videos 
  β”œβ”€β”€ videos/ # Dynamic real-world depth-map-based videos 
  └── dynamic/ # Dynamic synthetic motion videos 
      β”œβ”€β”€ images/ # Single object with dynamic motion 
      β”œβ”€β”€ words/ # Words with dynamic motion 
      └── dual_images/ # Two objects with dynamic motion

Metadata Schema

The metadata.csv file contains the following columns:

Column Description
file_name / Path Relative path to video file
Category Video category (images/words/shapes/videos/dual_images)
Label Object/word labels in the video (list format)
Noise Density Noise level from 10 to 90
Speckle Size Speckle size parameter (1, 3, etc.)
Frame Count Total number of frames
FPS Frames per second
Duration Video duration in seconds
Movement Type static or dynamic

Usage

Loading the Dataset

  from datasets import load_dataset
  
  # Load the entire dataset
  dataset = load_dataset("mukul54/spookybench-v2")
  
  # Access a video
  example = dataset["train"][0]
  video_path = example["file_name"]
  label = example["Label"]
  noise_level = example["Noise Density"]
  movement = example["Movement Type"]
  
  print(f"Video: {video_path}")
  print(f"Label: {label}")
  print(f"Noise: {noise_level}, Movement: {movement}")
  Filtering by Category
  # Filter static images only
  static_images = dataset.filter(
      lambda x: x["Category"] == "images" and x["Movement Type"] == "static"
  )
  
  # Filter dynamic videos only
  dynamic_videos = dataset.filter(
      lambda x: x["Movement Type"] == "dynamic"
  )
  
  # Filter high noise videos
  high_noise = dataset.filter(
      lambda x: x["Noise Density"] >= 70
  )

Example Use Cases

  • Noise Robustness Evaluation: Test model performance across different noise levels
  • Temporal Understanding: Compare model accuracy on static vs dynamic content
  • Zero-Shot Classification: Evaluate on diverse object categories in extreme cases
  • Security and Captcha Verification: This can provide an ideal captcha for websites.

Citation

If you use this dataset in your research, please cite us using bibtex:

  @article{upadhyay2025time,
    title={Time Blindness: Why Video-Language Models Can't See What Humans Can?},
    author={Upadhyay, Ujjwal and Ranjan, Mukul and Shen, Zhiqiang and Elhoseiny, Mohamed},
    journal={arXiv preprint arXiv:2505.24867},
    year={2025}
  }

Contact

For questions or issues regarding the dataset, please open an issue on the dataset repository or contact the authors.

Downloads last month
120