Automatic Hand Tracking with SAM

This project implements automatic hand tracking in videos using SAM2. It detects and masks hands in each frame of a video and outputs the processed video.

Features

• Hand Detection: Detects hand locations in video frames.
• Hand Masking: Generates segmentation masks for detected hands using SAM.
• Video Processing: Outputs a processed video with hands masked.

Prerequisites

• Python 3.8 or higher
• NVIDIA GPU with CUDA support (recommended for faster processing)

Installation

1. Clone the repository:

   git clone https://github.com/your-repo/automatic-hand-tracking.git
   cd automatic-hand-tracking

2. Install dependencies:

   pip install -r requirements.txt

3. Download the SAM checkpoint:

   wget -O sam_vit_b.pth https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth

Usage

1. Prepare your input video and place it in the project directory.

2. Run the script to process the video:

   python hand_tracking.py --input_path <input_video_path> --output_path <output_video_path>

   Replace <input_video_path> with the path to your input video and <output_video_path> with the desired output path.

3. The processed video with masked hands will be saved at the specified output location.

Dependencies

See requirements.txt for a list of dependencies.

Acknowledgements

This project utilizes Segment Anything Model (SAM) by Facebook Research.