got-10k / toolkit

Official Python toolkit for generic object tracking benchmark GOT-10k and beyond
http://got-10k.aitestunion.com/
MIT License
555 stars 95 forks source link

How to run a tracker on VOT2018 long-term benchmark. #28

Open 594422814 opened 4 years ago

594422814 commented 4 years ago

Thanks for releasing this great evaluation toolkit. I observe that this toolkit includes the VOT2018LT benchmark. I can successfully evaluated a tracker on the standard VOT2018 benchmark. However, I fail to evaluate the tracker on long-term benchmark using the same manner. Should I change the codes (e.g., the tracker definition)? Thanks!