yanghai-1218 / scflow Goto Github PK

Hi,

Thanks for your promising work, when I'm reproducing the code, there is an ImportError

ImportError: cannot import name 'build_augmentation' from 'models.utils' (/data_sata/WorkSpace_ZT/scflow/models/utils/init.py)

I can not find this module in utils, is there any missing file?

Hello, thank you very much for the code. May I ask in what way is Shape-Constraint embodied in which code exactly?

Dear Author,

Thank you very much for sharing the code to the community!

I tried to reproduce the YCBV results in your paper but failed. The first question came to my mind was "Can the provided pretrained ycbv_mixpbr reach around 0.82 AR of BOP metrics in YCBV dataset?" I can only reach around 0.79 AR.

Thanks a lot in advance!

Best,
Haixin

Thanks for sharing the great work.
I wonder how does the model works for novel objects that are not seen during training phase. Have you ever conducted similar experiments？

When I run python test.py, it reports an error: FileNotFoundError: RefineTestDataset: [Errno 2] No such file or directory: 'data/ycbv/keypoints/bbox.json'. However, where can I download the key points/bbox.json file. Because i didn't notice it in the bop data.

The open-source code does not contain the portion for replicating results on the LM and LMO datasets. Do you have any suggestions for reproducing these results?