Yolov5 advanced 7 target tracking latest environment setup

Previously, I introduced deepsort3.0 +yolo5.0 Target tracking environment construction process , In order to adapt to the new pt Environmental Science , Start trying to use more yolo6.0 The above version deepsort. Most of the new versions need to be completed by scientific Internet , You know .deepsort And its application instructions are shown in https://github.com/mikel-brostrom/Yolov5_StrongSORT_OSNet/blob/master/README.md
First establish the environment , Use 3.8 Or above ,

conda create --prefix=D:/PycharmProjects/DeepsortNew/DeepsortNew_env python=3.8

install pytorch

pip install torch==1.8.0+cu111 torchvision==0.9.0+cu111 torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html

take Yolov5_StrongSORT_OSNet-master(6.0 Later versions , Not yet release) Put in and env Parallel folders , And then yolov5-6.1（ When writing this article, I still master No, release） Change of name yolov Put it in Yolov5_StrongSORT_OSNet-6.0 Under the folder . Respectively into the yolov5 and Yolov5_StrongSORT_OSNet-master perform

pip install -r requirements.txt

Installation dependency . After commissioning yolov5
perform

python track.py --source 0

Tips
ModuleNotFoundError: No module named ‘torchreid’
Execute the following command to install the module

pip install https://github.com/KaiyangZhou/deep-person-reid/archive/master.zip

The test again yolov5 track
The execution will prompt
No URL associated to the chosen DeepSort weights. Choose between:
Available ReID models for automatic download
according to github readme Execute the following code （ Science and the Internet ）

python track.py --source 0 --strong-sort-weights osnet_x0_25_market1501.pt

Discover or HTTPS error , If you use it HTTPS The agent is successful . It can be downloaded automatically according to the prompt pt Include
resnet50_market1501.pt’, ‘resnet50_dukemtmcreid.pt’, ‘resnet50_msmt17.pt’, ‘resnet50_fc512_market1501.pt’, ‘resnet50_fc512_dukemtmcreid.pt’, ‘resnet50_fc512_msmt17.pt’, ‘mlfn_market1501.pt’, ‘mlfn_dukemtmcreid.pt’, ‘mlfn_msmt17.pt’, ‘hacnn_market1501.pt’, ‘hacnn_dukemtmcreid.pt’, ‘hacnn_msmt17.pt’, ‘mobilenetv2_x1_0_market1501.pt’, ‘mobilenetv2_x1_0_dukemtmcreid.pt’, ‘mobilenetv2_x1_0_msmt17.pt’, ‘mobilenetv2_x1_4_market1501.pt’, ‘mobilenetv2_x1_4_dukemtmcreid.pt’, ‘mobilenetv2_x1_4_msmt17.pt’, ‘osnet_x1_0_market1501.pt’, ‘osnet_x1_0_dukemtmcreid.pt’, ‘osnet_x1_0_msmt17.pt’, ‘osnet_x0_75_market1501.pt’, ‘osnet_x0_75_dukemtmcreid.pt’, ‘osnet_x0_75_msmt17.pt’, ‘osnet_x0_5_market1501.pt’, ‘osnet_x0_5_dukemtmcreid.pt’, ‘osnet_x0_5_msmt17.pt’, ‘osnet_x0_25_market1501.pt’, ‘osnet_x0_25_dukemtmcreid.pt’, ‘osnet_x0_25_msmt17.pt’, ‘osnet_ibn_x1_0_msmt17.pt’, 'osnet_ain_x1_0_msmt17.pt

choose
python track.py --source 0 --strong-sort-weight osnet_x0_75_msmt17.pt
success . You need to go online scientifically , It can be disconnected after the recognition is started
Be careful , If the prompt is missing
osnet_x0_75_imagenet.pth
Put this file Copied to the C:\Users\XXXX.cache\torch\checkpoints that will do , File download path
The file download address
https://download.csdn.net/download/buaaweibin/85840631

The suggestion still applies deepsort3.0 + yolov5 5.0 It's more convenient .