astra-vision / PaSCo

[CVPR 2024 Oral, Best Paper Award Candidate] Official repository of "PaSCo: Urban 3D Panoptic Scene Completion with Uncertainty Awareness"
https://astra-vision.github.io/PaSCo/
Apache License 2.0
149 stars 14 forks source link

Dataset input form #8

Closed Edric-star closed 2 months ago

Edric-star commented 2 months ago
/gpfsdswork/dataset/SemanticKITTI/dataset/
    └── sequences/
        └── 01/
            ├── labels/
            ├── velodyne/
            ├── voxels/
            ├── calib.txt
            ├── poses.txt
            └── times.txt

Hi there, should the dataset form goes like this? It's a little bit different from what's been posted in ReadMe file

anhquancao commented 2 months ago

Yes, you are right

Edric-star commented 2 months ago

Hello, I have successfully run 'eval.py' with semanticKITTI. When will you update the code to generate the completed scene files? I am eagerly looking forward to it! I would greatly appreciate it.

anhquancao commented 2 months ago

Hi, do you mean this section?

Edric-star commented 2 months ago

Hi, do you mean this section?

Sorry, I think this part seems to be about generating some panoramic labels, that is, files with added instance labels. I might still be learning and have some gaps in my understanding of this part. Could you please confirm if this is the final result of the scene generation?

Edric-star commented 2 months ago

Maybe I was talking about this part 5. Visualization in your ReadMe file as I just noticed, I would like to check the effect of the output completion files, just like these pics in your paper. Again, thank you for all your patience and replies, it really made my day. image

Edric-star commented 2 months ago

Thanks for the update

anhquancao commented 2 months ago

Hi @Edric-star, I added visualization code here.