Giters

hxwork / HandBooster_Pytorch

[CVPR 2024] HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions, Pytorch implementation.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Data annotation of Ho3D v2 dataset

M-leng opened this issue · comments

commented

Dear Hao Xu,

I have read your work on CVPR2023 and this work on CVPR2024.

I have one question of the data annotation of Ho3d v2 dataset.

In your work on cvpr2023, you noted that you used the pre-processed ground truths from handoccnet for ho3d dataset.

Did you also use the annotation of handoccnet on your new handbooster?

Looking forward for your reply.

Best wishes,
Zhiying

commented

Hi Zhiying,

Thanks for your interest. Yes, we use the pre-processed ground truths from HandOccNet to train both the diffusion model in our HandBooster and the reconstruction baselines.