YOLOv7 Object Tracking Using PyTorch, OpenCV and Sort Tracking
GPL-3.0 License
Clone the repository.
git clone https://github.com/RizwanMunawar/yolov7-object-tracking.git
Goto the cloned folder.
cd yolov7-object-tracking
Create a virtual envirnoment (Recommended, If you dont want to disturb python packages)
# Create the virtural envirnment
conda create -n yolov7objtracking python=3.10
# Activate the virtural envirnment
conda activate yolov7objtracking
# Create the virtural envirnment
python3 -m venv yolov7objtracking
# Activate the virtural envirnment
source yolov7objtracking/bin/activate
# Create the virtural envirnment
python3 -m venv yolov7objtracking
# Activate the virtural envirnment
cd yolov7objtracking
cd Scripts
activate
cd ..
cd ..
Update pip and install libraries
# Upgrade pip with mentioned command below.
pip install --upgrade pip
# Install requirements with mentioned command below.
pip install -r requirements.txt
Run the script
Select the appropirate command from the following list of command according to your need. (by default, pretrained yolov7 weights will be automatically downloaded into the working directory if they don't already exist).
# for detection only
python detect.py --weights yolov7.pt --source "your video.mp4"
#if you want to change source file
python detect_and_track.py --weights yolov7.pt --source "your video.mp4"
#for WebCam
python detect_and_track.py --weights yolov7.pt --source 0
#for External Camera
python detect_and_track.py --weights yolov7.pt --source 1
#For LiveStream (Ip Stream URL Format i.e "rtsp://username:pass@ipaddress:portno/video/video.amp")
python detect_and_track.py --source "your IP Camera Stream URL" --device 0
#for specific class (person)
python detect_and_track.py --weights yolov7.pt --source "your video.mp4" --classes 0
#for colored tracks
python detect_and_track.py --weights yolov7.pt --source "your video.mp4" --colored-trk
#for saving tracks centroid, track id and bbox coordinates
python detect_and_track.py --weights yolov7.pt --source "your video.mp4" --save-txt --save-bbox-dim
Output file will be created in the working-dir/runs/detect/obj-tracking
with original filename.
For more details, you can reach out to me on Medium or can connect with me on LinkedIn