Using YOLOv3 and DEEPSORT, this project attempts to track multiple objects on a screen and assign them a unique id to reduce overcounting. This project return a cv2 screen with the classifications and also prints out the FPS and classes detected.
- Python --3.7.6
- Conda --4.8.3
git clone https://github.com/nakul-shahdadpuri/trackid.git
cd trackid/
conda env create -f setup.yml
conda activate trackid
pip install -r setup.txt
python run.py
- Non Max Suppression 'https://towardsdatascience.com/non-maximum-suppression-nms-93ce178e177c'
- YOLOv3 model 'https://pjreddie.com/darknet/yolo/'
- cv2.BlobFromImage 'https://www.pyimagesearch.com/2017/11/06/deep-learning-opencvs-blobfromimage-works/'
- OpenCv Documentation 'https://docs.opencv.org/2.4/'
- DeepSort Repo 'https://github.com/nwojke/deep_sort'
- SORT Paper 'https://arxiv.org/abs/1602.00763'
- Deep Sort 'https://medium.com/analytics-vidhya/yolo-v3-real-time-object-tracking-with-deep-sort-4cb1294c127f'