search

fabro66 / GAST-Net-3DPoseEstimation

A Graph Attention Spatio-temporal Convolutional Networks for 3D Human Pose Estimation in Video (GAST-Net)
MIT License
312 stars 70 forks source link

how the change the "num_joints_out" to the any number different with "num_joints_in" #38

Open we1314love opened 3 years ago

we1314love commented 3 years ago

@fabro66 thank you for your great work,

I have a question about the "num_joints_out -- number of output joints (can be different than input)" in gast_net.py, generally the "num_joints_in" is equal to "num_joints_out", but if I will to use the whole 17 joints of 2D pose to predict 3(any number unequal to 17) joints of 3D pose, do you have any suggestions how to do that?

fabro66 commented 3 years ago

Hi～

You can use up-sampling or down-sampling to reduce or increase the number of output joint points.

Reference: Exploiting-Spatial-temporal-Relationships-for-3D-Pose-Estimation-via-Graph-Convolutional-Networks