search

CMU-Perceptual-Computing-Lab / openpose

OpenPose: Real-time multi-person keypoint detection library for body, face, hands, and foot estimation
https://cmu-perceptual-computing-lab.github.io/openpose
Other
31.02k stars 7.85k forks source link

About single person detection #1964

Open soheilAppear opened 3 years ago

soheilAppear commented 3 years ago

Thank you for your great works! I have a question regarding the algorithm for singular detection. I saw you have used the bottom-up approach since it has a higher speed for multiple-person detection. My question is does this persons' keypoints detection is getting done during running the model or it is getting done during the post-processing part in which we run the inference?

I'm trying to speed up the algorithm for a single person since I only want to detect one person at a time without extra calculation for other multiple people in each frame.

Chauncey-Wang commented 3 years ago

You can set params["number_people_max"] = 1

stale[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.