MVIG-SJTU / AlphaPose

Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System
http://mvig.org/research/alphapose.html
Other
7.86k stars 1.95k forks source link

想只识别图片中最主要的那一个人物的姿态估计怎么做 #1083

Open zzzwwb opened 1 year ago

zzzwwb commented 1 year ago

您好,我想在图片中只识别一个人的姿态骨骼关键点,应该怎么做啊? 比如花样滑冰里不被背景里的观众影响

timbsb commented 11 months ago

You can try that using the human with the biggest area BBOX.