search

CVMI-Lab / DODA

(ECCV 2022) DODA: Data-oriented Sim-to-Real Domain Adaptation for 3D Semantic Segmentation
Apache License 2.0
46 stars 4 forks source link

3D front dataset preprocessing script #5

Closed msingh27 closed 2 years ago

msingh27 commented 2 years ago

Hi,

Thanks a lot for open-sourcing your exciting work!

I was wondering if you can share the preprocessing script (point cloud sampling + label assignment script) for 3dFront?

Is there a way to obtain instance id information (along with semantic category) for each point in the point cloud in the processed dataset "density1250.zip".

It would be great if you could explain the meaning of the values in points[:, 3:6] as mentioned in code below https://github.com/CVMI-Lab/DODA/blob/b5f694202cdff94c6e013b2b5b1bc8ddfd22f0c9/dataset/front3d.py#L46

Dingry commented 2 years ago

Hi, you can refer to the this link for 3D-FRONT processing: https://github.com/3D-FRONT-FUTURE/3D-FRONT-ToolBox/tree/master. The point cloud sampling is based on CloudCompare with density 1250. The label taxtonomy is provided in the suppl. file. We'll make the pre-processing scripts public once we've organized it.

points[:, 3:6] denotes the coordinations, which is stored in the column 0-3 in points array.

CurryYuan commented 1 year ago

Hi @Dingry ,

Is there any progress on the preprocessing script?