This project demonstrates how to detect people in a video using the YOLO (You Only Look Once) object detection model with OpenCV in Python. The code processes each frame of the video to identify and highlight people with bounding boxes.
- Python 3.x
- OpenCV
-
Clone the Repository
git clone https://github.com/Dev-wael1/video-person-detection.git cd video-person-detection -
Install Dependencies
Install the required Python libraries using pip:
pip install opencv-python opencv-python-headless
-
Download YOLO Files
Download the YOLOv3 weights and configuration files from the YOLO website. Place these files in the project directory:
yolov3.weightsyolov3.cfg
You may also need the
coco.namesfile which contains the class labels. You can download it from here.
-
Prepare Your Video
Place your video file in the project directory and rename it to
video.mp4, or update the script to use your video's filename. -
Run the Detection Script
Execute the Python script to start processing the video:
python detect_people.py
The script will open a window displaying the video with detected people highlighted. Press
qto exit the video window.
- Loading YOLO: The script loads the YOLO model using pre-trained weights and configuration files.
- Processing Video: It reads frames from the video, applies YOLO object detection, and draws bounding boxes around detected people.
- Non-max Suppression: Reduces overlapping bounding boxes to ensure each detected person is only marked once.
- Ensure that all required files (
yolov3.weights,yolov3.cfg,coco.names, and the video file) are in the correct directory. - Adjust the confidence threshold and non-max suppression parameters in the script as needed for better results.
This project is licensed under the MIT License - see the LICENSE file for details.
- YOLO: https://pjreddie.com/darknet/yolo/
- OpenCV: https://opencv.org/