Error while setting up environment & RDIF training code

[anticdimi]

Hi @Daisy-1227 ,

While trying to set up the environment, the command pip install git+https://github.com/zilch/pyigl outputs an error: fatal: repository 'https://github.com/zilch/pyigl/' not found. Indeed, this repository doesn't exist.

Could you point me to the correct one which should be used?

[Daisy-1227]

Hi @anticdimi,

After checking, I found that there was an error in this URL. It should actually be 'https://github.com/zzilch/pyigl' instead of 'https://github.com/zilch/pyigl'. Thank you for the reminder, and I will update the README.

[anticdimi]

Thanks for the prompt reaction!
I will close this issue 😄

[anticdimi]

Actually, while I'm here, I would like to ask one more question. 🥲

Do you plan to release the code to train the RDIF model? Or is it already here on the repo and I missed it?
I can't seem to find any instructions in README..

[Daisy-1227]

@anticdimi

The training of the RDIF model was done by Zeyu Huang, who has graduated.

If you need further details, you can refer to the training details of DIF. Also, I will try to organize and release this code when I have some free time (I am currently busy with another project 😂).

[anticdimi]

Oh okay. If you have time at some point, it'd be great.
But given what you write in supplementary materials and main paper, the training should be exactly as the one of DIF-Net; so theoretically, I could just re-use their training loop? With the "extension" of randomly rotating the meshes as "augmentation" needed for $E_R, E_\alpha$?

Feel free to close this issue, I will follow the repository for updates regarding this.