search

fabro66 / GAST-Net-3DPoseEstimation

A Graph Attention Spatio-temporal Convolutional Networks for 3D Human Pose Estimation in Video (GAST-Net)
MIT License
314 stars 69 forks source link

Multi-person real-time inference #69

Open Ared521 opened 1 year ago

Ared521 commented 1 year ago

Hello, thank you for your great work. Has the real-time attitude estimation of multiple people been completed?

sk-zhang commented 1 year ago

Hi, have you already realized it?

Ared521 commented 1 year ago

Hi, have you already realized it?

No, i think 2d sequence to 3d pose is diffcult in multi-person.

sk-zhang commented 1 year ago

Hi, have you already realized it?

No, i think 2d sequence to 3d pose is diffcult in multi-person.

Thanks for the reply.