Ali2500/BURST-benchmark

Training setting

z-jiaming opened this issue · 1 comments

z-jiaming commented

Hi, thanks for your work.
I want to know the training setting of your STCN baseline in mask-based Exemplar Guided, i.e., 49.8 HOTA_all?
Is it just trained on BURST training set, or just using off-the-shelf STCN model trained on DAVIS&YouTube-VOS?

Thanks!!

Ali2500 commented

The STCN baseline was generated using off-the-shelf weights trained by the authors on DAVIS+YouTubeVOS