how to do layer by layer inference.

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

Open vicentowang opened 1 year ago

vicentowang commented 1 year ago

FPS of layer by layer inference is amazing. I wonder how to do that. Thanks