Ruizhi Cheng

Hello, I would like to ask could I use this framework to get SMPL-X model from multiple RGB-D sensors? My setup is: I will have 3-4 RGB-D sensors around the...

Any pytorch implementation available

1

Hello, I would like to ask is there any pytorch version of this project available. I found one: https://github.com/square-1111/3D-Point-Cloud-Modeling. But it is not an official implementation and not trained. Moreover,...

Why the output mesh is already in the origin?

2

Hi, when I visualize the output mesh using meshlab, I found the mesh has already been in the origin, but not in the position that has "transl" towards camera's position...

Just feed translation when initialization camera and put camera.translation.requires_grad = False?

multiview SMPLX

2

Hello, I would like to ask whether you will release the multi-view version of SMPLify-X you used for SMPLX annotation mentioned in section 3.4?

I was trying to use this work to estimate my pose. I load the beat value obtained from [smplify-x](https://github.com/vchoutas/smplify-x). However, I found the pose of output mesh is totally wrong,...

Hi, thanks for the great work! I would like to ask if I want to use 3D pose as the input, how should I re-train the model? Basically, I want...

How can we generate human movements over time

1

Thank you for your great work! After reading the paper, one question I have is that it seems the current PoseGPT can work only on a single-frame basis and cannot...

Inference time per frame

2

Hi, thank you for the awesome work! I would like to ask what is the inference time of it? Can it achieve real-time performance?