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 modified the code according to your method 1(https://github.com/CMU-Perceptual-Computing-Lab/MonocularTotalCapture/issues/20). It can run successfully on the remote server, but the output result is very poor. It seems that render is necessary for optical flow. How can I modify it to run correctly on the remote server? I look forward to your reply
The text was updated successfully, but these errors were encountered:
Hi, I also encountered this problem.
I have tried to save the fitting time by commending the generateMesh and some functions from meshTrackingProj.h to not save the rendered mesh png results since the saved txt contains the model parameters, but the tracking results become worse than the untracked results. and I also tried to commend the lines that contain render, but even the txt don't save.
Hope you guys can tell how to not visualize and not save the rendered mesh, thanks~
I modified the code according to your method 1(https://github.com/CMU-Perceptual-Computing-Lab/MonocularTotalCapture/issues/20). It can run successfully on the remote server, but the output result is very poor. It seems that render is necessary for optical flow. How can I modify it to run correctly on the remote server? I look forward to your reply
The text was updated successfully, but these errors were encountered: