search

hongsukchoi / Pose2Mesh_RELEASE

Official Pytorch implementation of "Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose", ECCV 2020
MIT License
665 stars 69 forks source link

How fast can the inference time be achieved? #31

Open favorxin opened 2 years ago

favorxin commented 2 years ago

Hello! This work is very meaningful! I want to inference the video in 720P. How much can FPS achieve?

hongsukchoi commented 2 years ago

Hi,

Pose2Mesh itself achieves 37FPS, given the input 2D pose. Check out the FPS of 2D pose estimators. Also, refer to Section 5 implementation details of the paper.

favorxin commented 2 years ago

Hi,

Pose2Mesh itself achieves 37FPS, given the input 2D pose. Check out the FPS of 2D pose estimators. Also, refer to Section 5 implementation details of the paper.

Thanks for quick replying! I try to run the demo/run.py to inference my pics. But I went the following errors. image

hongsukchoi commented 2 years ago

Will you make a 'result' directory under 'demo' directory?