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
I think we can get great end to end speedup if we convert make_bv_feature function to pytorch gpu. This is the most time consuming function right now as it is done using numpy. Have you thought about it?
Hi,
Great work.
May I know the inference speed in ms or in FPS?
Thanks.
Deepak
The text was updated successfully, but these errors were encountered: