|
|
--- |
|
|
title: Glad8tr Video Analysis |
|
|
emoji: 🎥 |
|
|
colorFrom: blue |
|
|
colorTo: indigo |
|
|
sdk: gradio |
|
|
sdk_version: 4.0.0 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
# Glad8tr Video Analysis |
|
|
|
|
|
This is a video analysis application that uses computer vision to detect objects, analyze poses, and provide cognitive state analysis. The application is deployed on Hugging Face Spaces. |
|
|
|
|
|
## Features |
|
|
|
|
|
- Object detection and tracking |
|
|
- Pose estimation and analysis |
|
|
- Scene context analysis |
|
|
- Cognitive state analysis |
|
|
- Real-time video processing |
|
|
|
|
|
## Usage |
|
|
|
|
|
1. Upload a video file (supported formats: MP4, AVI, MOV) |
|
|
2. The application will process the video and provide: |
|
|
- Object detection results |
|
|
- Pose analysis |
|
|
- Scene context information |
|
|
- Cognitive state analysis |
|
|
3. Download the processed video with annotations |
|
|
|
|
|
## Technical Details |
|
|
|
|
|
- Built with PyTorch and YOLOv8 |
|
|
- Uses Gradio for the web interface |
|
|
- Optimized for Hugging Face Spaces deployment |
|
|
- Processes videos in real-time with frame sampling |
|
|
|
|
|
## Limitations |
|
|
|
|
|
- Processing is limited to the first 100 frames for demo purposes |
|
|
- Maximum video resolution: 1920x1080 |
|
|
- Processing time depends on video length and complexity |
|
|
|
|
|
## Model Information |
|
|
|
|
|
The application uses two YOLOv8 models: |
|
|
1. Object detection model (glad8trv8s.pt) |
|
|
2. Pose estimation model (glad8trv8s-pose.pt) |
|
|
|
|
|
If the custom models are not available, the application will fall back to the base YOLOv8 models. |
|
|
|
|
|
## License |
|
|
|
|
|
This project is licensed under the MIT License. |
