iglict / ibsr_jittor Goto Github PK

Hi, thanks for your sharing. But there are some files not contained in this repo, such as rendering_pix3d.pkl, pix3d_train.json, pix3d_test.json and pretrained models. Can you submit these files if possible? Thanks!

Hi~ @tommaoer, thank you for the repo.
As shown in (https://github.com/IGLICT/IBSR_jittor/blob/main/code/RetrievalNet.py#L3), 'from Models import RetrievalNet'
but I can not find the file, where is the Models.py and RetrievalNet?

Hello!
Could you please provide us with the train / test splits that you have used in your paper for the Pix3D dataset?
Also, I have noticed that in your paper you say that you are using 94 3D shapes from Comp Cars 3D, whereas the original Comp Cars 3D repository has 98 3D shapes (annotations are here). Is there a specific reason for removing 4 3D shapes?

Thanks a lot!
Konstantinos

Hello, thanks for sharing your work!

As I'm trying to reproduce your results on Pix3D, could you provide important data files including rendering_pix3d.pkl, pix3d_train.json, pix3d_test.json, predicted masks from OCRNet and so on.
It would be very helpful if you can provide more details about how you train OCRNet on Pix3D and pretrained OCRNet model.

Thanks!

Hello, Thank you very much for your work, Where can I find the pre-trained models?