This repository implements YOLO (You Only Look Once) object detection algorithm tailored for video analysis. YOLO is a state-of-the-art, real-time object detection system that can detect multiple objects in an image or frame in one pass.
- Video Object Detection: Detect objects in videos using the YOLO algorithm.
- Real-Time Processing: Fast and efficient processing for real-time or near real-time applications.
- Pre-trained Models: Support for multiple pre-trained YOLO models for various object detection tasks.
- Easy Integration: Simple interface for integrating YOLO into video processing pipelines.
-
Clone the repository:
git clone https://github.com/anjanbudige/yolo-object-detection.git
-
Install dependencies:
use colab or anaconda to open python notebook file (pynb file)
git clone https://github.com/WongKinYiu/yolov9 cd yolov9 pip install -r requirements.txt
-
Download pre-trained YOLO weights and configuration files. Below we have given pretrained models, you can use them for object detection
To perform object detection on a video, follow these steps:
-
Ensure dependencies are installed and pre-trained model files are downloaded.
-
Run the detection script:
python detect_video.py --weights "model.pt file" --conf "0.1 to 1.0" --source path/to/input/video.mp4 --device cpu
- Replace
path/to/input/video.mp4
with the path to your input video file. - Provide paths to YOLO weights, configuration, and class files.
- Replace
-
Adjust parameters as needed.
This repository includes links to pre-trained YOLO weights and configuration files for various YOLO versions:
- YOLOV9 c Converted: Download
- YOLOV9 e Converted: Download
- YOLOV9 c: Download
- YOLOV9 e: Download
- Gelan c: Download
- Gelan e: Download
Contributions are welcome! If you have any ideas, enhancements, or bug fixes, feel free to open an issue or submit a pull request.
This project is licensed under the MIT License - see the LICENSE file for details.