zc-alexfan/hold
[CVPR 2024✨Highlight] Official repository for HOLD, the first method that jointly reconstructs articulated hands and objects from monocular videos without assuming a pre-scanned object template and 3D hand-object training data.
PythonMIT
Issues
- 1
What version of pip are you using
#22 opened by shmooWang - 3
Evaluation server submission
#19 opened by zc-alexfan - 2
- 3
Problems on the optimize_ckpt.py
#20 opened by ACondaway - 0
Inconsistency in HO3D Data's Prediction
#17 opened by alakhag - 6
KeyError while reading misc to optimize model
#16 opened by Salingo - 3
Large dynamic interaction examples
#8 opened by anilesec - 5
Problems on ECCV'2024 Workshop on ARCTIC
#11 opened by ACondaway - 2
Training on Mutiple GPUs
#15 opened by eccv-workspace - 1
problem in setup metro env
#12 opened by gemcollector - 1
- 5
problem on getting total mesh
#13 opened by Lyy-iiis - 7
- 1
How to train from a single video
#10 opened by Lyy-iiis - 6
Submodule not found
#9 opened by KaanKarakoseELTE - 3
Code release
#7 opened by tunantu - 10
Code Release
#1 opened by anilesec - 0
General tips on better reconstruction
#6 opened by zc-alexfan - 2
- 3
Training dataset
#4 opened by yilan120 - 1
How many frames do you used?
#2 opened by EAST-J