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
Hello,
Thank you for your amazing work, I was wondering since the inference times are so low, would it be possible to use this model in real-time with the input being my webcam for example ? If so, what do you recommend doing to optimize ?
Thank you so much
The text was updated successfully, but these errors were encountered:
Thank you for your interest! The inference is actually very fast if you compute the per-frame processing time (note that our method processes a graph at a time, which includes many frames). And yes, it's possible to extend our work in real time. We are currently working on this problem and will release it within this year! Please stay tuned.
Hello,
Thank you for your amazing work, I was wondering since the inference times are so low, would it be possible to use this model in real-time with the input being my webcam for example ? If so, what do you recommend doing to optimize ?
Thank you so much
The text was updated successfully, but these errors were encountered: