search

jiawen-zhu / ViPT

[CVPR23] Visual Prompt Multi-Modal Tracking
MIT License
256 stars 18 forks source link

How to test on RGB-D dataset by vot-toolkit？ #17

Open hoqolo opened 1 year ago

hoqolo commented 1 year ago

Your work is very attractive to me. I was studying your work recently, but encountered a difficult problem when testing and spent several days without solving it. May I ask how to test the RGB-D related experiments? I use VOT-toolkit to test according to your README file, but there are many errors, such as "Unable to resolve tracker reference: deep_rgbd", “Tracker deep_rgbd encountered an error: Unable to connect to tracker ”, please give the detailed test bash code and related precautions? Looking forward to your reply.

image image

hoqolo commented 1 year ago

I found that it seems to be a problem with the package version, what is the version of your vot-toolkit and vot-trax? If possible, can you publish your environment configuration, thank you!

jiawen-zhu commented 1 year ago

Hi~ We employ the vot-toolkit==0.5.3 with vot-trax==3.0.3

hoqolo commented 1 year ago

This works, thank you very much. Great job!

hyx1110 commented 11 months ago

After I execute this command vot evaluate --workspace ./ vipt_deep, I encountered the following error. How can I solve this problem? Looking forward to your reply. fba3a30baa70ea9871312c37784ee48

923714892 commented 1 month ago

After I execute this command vot evaluate --workspace ./ vipt_deep, I encountered the following error. How can I solve this problem? Looking forward to your reply.

I solve this problem by using vot-trax=4.0.2 vot-toolkit=0.7.1 and replace the vot.py with a new one in the official website of vot-trax.Hope to help you.