Zhongdao / UniTrack

[NeurIPS'21] Unified tracking framework with a single appearance model. It supports Single Object Tracking (SOT), Video Object Segmentation (VOS), Multi-Object Tracking (MOT), Multi-Object Tracking and Segmentation (MOTS), Pose Tracking, Video Instance Segmentation (VIS), and class-agnostic MOT (e.g. TAO dataset).
MIT License
338 stars 34 forks source link

bad result #27

Open liuyang77712 opened 2 years ago

liuyang77712 commented 2 years ago

hi @Zhongdao , I have a problem. I had a bad result like this pic. 00162

I mean it should track more cars in pic.

I only donwload the yolox_x.pth and run the python demo. what should i do?

Zhongdao commented 2 years ago

Hi @liuyang77712 This is a bug in a previous version of the code, have you tried the newest version?

liuyang77712 commented 2 years ago

Hi @liuyang77712 This is a bug in a previous version of the code, have you tried the newest version?

Hi @Zhongdao I retry this project ,there are some problems. 1.git clone . there are some warnings: like 373e6ce3c60956dfc007b485d3a546e 2.install pytorch:conda install pytorch==1.7.0 torchvision cudatoolkit 7e783e84fc824b2864c78f9d0991c30 so.I go to pytorch.org to download it 3.I go to detector/yolox to install requirements.txt , when I use this command:setup.py . there is a problem 8dfb5bd24ffd1262d40d952b0f326db I change the code like 91c832c168fb31739c014e8657e2c89 finally 1be1835054ad7f3e24e7490ec79cc2b

and I use the newest code.

what should i do?