search

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
334 stars 35 forks source link

How to retrain you model ? #41

Open CarlHuangNuc opened 1 year ago

CarlHuangNuc commented 1 year ago

Hi, i want to retrain your model in custom dataset ?

How to train your model, can you give a guide ?

Zhongdao commented 1 year ago

Hi @CarlHuangNuc, In UniTrack we do not perform training on tracking datasets, instead, we adopt pre-trained models directly for the tasks.
You can try any pre-trained models, train on any datasets, and test with UniTrack.