This project implements object detection using YOLOv3 with pre-trained weights. It supports live detection from a webcam, image detection, and video detection. The application is built using Python with libraries such as OpenCV, PIL, and Tkinter for the GUI, and runs primarily through a Jupyter Notebook interface.
- Live Detection: Real-time object detection using a webcam.
- Image Detection: Detect objects in images.
- Video Detection: Detect objects in video files.
- Save Results: Save the processed images and videos with detected objects.
- Python: Core programming language.
- OpenCV: For image and video processing.
- PIL (Pillow): For image handling.
- Tkinter: For the graphical user interface.
- YOLOv3: Object detection model.
- Jupyter Notebook: For running and displaying the main code.
- Clone the repository:
git clone https://github.com/yourusername/object-detection-yolov3.git cd object-detection-yolov3
- Install dependencies:
- Make sure you have Python installed. Then, install the required libraries:
pip install numpy opencv-python pillow matplotlib jupyter
- Download YOLOv3 weights and configuration:
- Download the YOLOv3 weights from the official source. (https://pjreddie.com/darknet/yolo/)
- Download the yolov3.cfg and coco.names files. (already present in repository)
- Set up the project:
- Place yolov3.weights, yolov3.cfg, and coco.names in the project directory.
- Run Jupyter Notebook:
jupyter notebook
Open the object_detection.ipynb notebook.
- Live Detection:
- Click on live detection to start real-time object detection using your webcam.
- Image Detection:
- Click on image detection, select an image file, and the application will detect objects in the image.
- Video Detection:
- Click on video detection and the application will detect objects in the video.