Yolo-R DeepSort

• This repository was built by combining Yolo-R object detection and Deepsort algorithm.

Installation

1. First, clone the repository.

git clone https://github.com/chunmusic/Yolo-R_DeepSort

2. Create Python virtual environment based on Python 3.8

conda create -n py38 python=3.8

3. Activate Python 3.8 virtual environment

conda activate py38

4. Install required dependencies.

conda install pytorch torchvision cudatoolkit=11.1 -c pytorch -c conda-forge

pip install Cython numpy

pip install -r requirement.txt

5. Download yolor weight file (.pt) from yolor_p6.pt

How to use

Inference

• Run the following command to test inference with video file.

python detect.py --source walk.mp4 --cfg cfg/yolor_p6.cfg --weights yolor_p6.pt --conf-thres 0.25 --img-size 1280 --device 0 --view-img

Screenshot

Reference

• yolor
• Yolov5_DeepSort_Pytorch
• deep_sort