search

researchmm / TracKit

[ECCV'20] Ocean: Object-aware Anchor-Free Tracking
MIT License
613 stars 97 forks source link

Which datasets are supported for eval ？ #44

Closed xiaofeinot closed 3 years ago

xiaofeinot commented 4 years ago

If I use "python lib/eval_toolkit/bin/eval.py " to eval the performance of model, which datasets are supported? Only VOT or some of VOT, OTB, GOT10K, LASOT ? Thanks !

JudasDie commented 4 years ago

If I use "python lib/eval_toolkit/bin/eval.py " to eval the performance of model, which datasets are supported? Only VOT or some of VOT, OTB, GOT10K, LASOT ? Thanks !

Hi, VOT is supported. You can change val benchmark referring to VOT codes.

xiaofeinot commented 4 years ago

If I use "python lib/eval_toolkit/bin/eval.py " to eval the performance of model, which datasets are supported? Only VOT or some of VOT, OTB, GOT10K, LASOT ? Thanks !

Hi, VOT is supported. You can change val benchmark referring to VOT codes.

Hi, so that's mean that if I want to eval the tracker in other datasets, I have to refer the VOT codes and write codes myself? Is it right?

JudasDie commented 4 years ago

If I use "python lib/eval_toolkit/bin/eval.py " to eval the performance of model, which datasets are supported? Only VOT or some of VOT, OTB, GOT10K, LASOT ? Thanks !

Hi, VOT is supported. You can change val benchmark referring to VOT codes.

Hi, so that's mean that if I want to eval the tracker in other datasets, I have to refer the VOT codes and write codes myself? Is it right?

Yes. In fact, only small changes are expected.