Train module

[neneyork]

Thanks for your work in monocular depth estimation of endocsope. I would like to test the performance of this nerwork on endocsope videos with surgical tools and shadows. Could you please release the code of training module? Thank you!

[yahskapar]

Hi @neneyork,

Thanks for taking an interest in our work! For the time being, I've only released our pre-trained models and inference code (suited for C3VD, but easily adaptable for other datasets). Please keep an eye on this repo (star it and watch it) for the training code in the near future - my current plan is to have it fully cleaned and released by late May or so.

If you need help in the meantime adapting the inference code for other datasets, or have any other questions, please let me know.

Best regards,

Akshay

EDIT: I've decided to delay releasing the full training code because this work itself is undergoing some changes (after being in submission at a conference) and I would like to wait until those changes are finalized.