Yibo Xia

icon indicating a website link https://www.buaa.edu.cn/

icon company Beihang University icon location 37 Xueyuan Road, Haidian District, Beijing, P.R. China, 100191. icon location Beihang University,major in CS

Results 7 comments of Yibo Xia

About training code

Agree, I also want to follow this great work and train the PoseVAE on my datasets. Looking forward to your release!

AttributeError: _3D

change to THREE_D

Inverting the new_joint_vecs?

> Hi, @andrewnc > > We actually do not use rotation in our representation from our experiments in the appendix. And we followed Humanml3D format to visualize the motion using...

convert the tomato representation to SMPL-X format for visualization.

> you can refer to these code: https://github.com/OpenMotionLab/MotionGPT/blob/main/fit.py, https://github.com/setarehc/diffusion-motion-inbetweening/blob/main/visualize/vis_utils.py > > very very strange / > Their representation is based on HumanML3D, which I think is very odd and strange....

About the training loss of Masked Transformer and Residual Transformer.

> Have you addressed this problem or is this issue acceptable? @bob35buaa > > Thanks. No, the author hasn't reply this question.

About the training loss of Masked Transformer and Residual Transformer.

> Sorry for the late response. I just found some time to verify the model on KIT. The training curve and accuracy are normal given the limited size of KIT....

Question about SignVAE

> In the paper, you propose SignVAE. I want to make 3D sign language avatar from the input text. Can you publish SignVAE Code? I have the same question @ZhengdiYu...