You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
To test our tracker on a custom dataset, I recommend starting with having a look and adapting the following files for your needs:
To generate the data, you can adapt the prepare_eds_subseq.py to obtain the timesurfaces required for the tracker. You can also experiment with the duration of the time window (dt) to achieve better performance. To use our data loaders, ensure that you have a similar data structure as the proposed training data.
To run the tracker, I recommend adapting the evaluate_real.py file. If you use the tracker in our framework (same data-loading, config files, etc.), this should be straight forward.
Can I use a custom dataset, such as the BS-ERGB dataset, for inference?
The text was updated successfully, but these errors were encountered: