YanjieZe/3D-Diffusion-Policy
[RSS 2024] 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations
PythonMIT
Issues
- 1
_copy_to_cpu not defined
#35 opened - 3
prediction_type for noise_scheduler
#34 opened - 2
- 10
sim env for realdex_drill dataset
#32 opened - 1
- 5
Question about run in simulate env
#30 opened - 2
Why we must to use MultiStepWrapper?
#29 opened - 1
onedrive link
#28 opened - 2
End-effector state class imablance
#26 opened - 2
- 4
Calibration and normalization
#24 opened - 1
Clarification on the meaning of action?
#23 opened - 3
Reproduce results
#22 opened - 2
About dexart benchmark
#21 opened - 1
- 5
about experiment
#19 opened - 10
How do I fix it?
#17 opened - 3
data size about train
#16 opened - 2
Question about the experiment in paper
#15 opened - 3
Data
#14 opened - 2
Why "downsample points by farthest point sampling" instead of other methods (like voxel filter)?
#13 opened - 1
Does data need to be in world frame?
#12 opened - 2
Meta-World benchmark code
#11 opened - 2
- 1
- 2
isaacgym example
#7 opened - 2
Something about training?
#6 opened - 12
- 2
Image to 3D point cloud?
#4 opened - 3
Training on real data?
#3 opened - 2
MuJoCo compilation issues
#2 opened - 3
About experiment
#1 opened