Yolov5 + Deep Sort with OSNet

Introduction

This repository contains a highly configurable two-stage-tracker that adjusts to different deployment scenarios. The detections generated by YOLOv5, a family of object detection architectures and models pretrained on the COCO dataset, are passed to a Deep Sort algorithm which combines motion and appearance information based on OSNet in order to tracks the objects. It can track any object that your Yolov5 model was trained to detect.

Tutorials

• Yolov5 training on Custom Data (link to external repository) 
• DeepSort deep descriptor training (link to external repository) 
• Yolov5 deep_sort pytorch evaluation 

Before you run the tracker

1. Clone the repository recursively:

git clone --recurse-submodules https://github.com/mikel-brostrom/Yolov5_DeepSort_Pytorch.git

If you already cloned and forgot to use --recurse-submodules you can run git submodule update --init

2. Make sure that you fulfill all the requirements: Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install, run:

pip install -r requirements.txt

Tracking sources

Tracking can be run on most video formats

$ python track.py --source 0  # webcam
                           img.jpg  # image
                           vid.mp4  # video
                           path/  # directory
                           path/*.jpg  # glob
                           'https://youtu.be/Zgi9g1ksQHc'  # YouTube
                           'rtsp://example.com/media.mp4'  # RTSP, RTMP, HTTP stream

Select object detection and ReID model

Yolov5

There is a clear trade-off between model inference speed and accuracy. In order to make it possible to fulfill your inference speed/accuracy needs you can select a Yolov5 family model for automatic download

$ python track.py --source 0 --yolo_model yolov5n.pt --img 640
                                          yolov5s.pt
                                          yolov5m.pt
                                          yolov5l.pt 
                                          yolov5x.pt --img 1280
                                          ...

DeepSort

The above applies to DeepSort models as well. Choose a ReID model based on your needs from this ReID model zoo

$ python track.py --source 0 --deep_sort_model osnet_x0_5_market1501
                                               resnet50_MSMT17
                                               mobilenetv2_x1_4_dukemtmcreid
                                               ...

Filter tracked classes

By default the tracker tracks all MS COCO classes.

If you only want to track persons I recommend you to get these weights for increased performance

python3 track.py --source 0 --yolo_model yolov5/weights/crowdhuman_yolov5m.pt --classes 0  # tracks persons, only

If you want to track a subset of the MS COCO classes, add their corresponding index after the classes flag

python3 track.py --source 0 --yolo_model yolov5s.pt --classes 16 17  # tracks cats and dogs, only

Here is a list of all the possible objects that a Yolov5 model trained on MS COCO can detect. Notice that the indexing for the classes in this repo starts at zero.

MOT compliant results

Can be saved to your experiment folder runs/track/<yolo_model>_<deep_sort_model>/ by

python3 track.py --source ... --save-txt

Cite

If you find this project useful in your research, please consider cite:

@misc{yolov5-deepsort-osnet-2022,
    title={Real-time multi-object tracker using YOLOv5 and deep sort with OSNet},
    author={Mikel Broström},
    howpublished = {\url{https://github.com/mikel-brostrom/Yolov5_DeepSort_OSNet}},
    year={2022}
}