Yuanchen Guo

I ran into the same problem with PyTorch 1.8.2 + tinycudann >= 1.5. It works fine with tinycudann 1.4. I solved the problem by upgrading PyTorch to the latest version...

Sure! You can add the ProlificDreamer config to this list: https://github.com/threestudio-project/threestudio/blob/38f50c1ad37e95d2ec67133bdbb4b683042f26d6/gradio_app.py#L77-L84 The first item in the tuple is the displayed name in the dropdown menu, and the second item is...

Thank you @07hyx06 ! I plan to add geometry quality comparison after I finish working on the neus-background branch :)

It seems that the white color results from the changing exposure / the use of flashlight during capture. You may refer to [this comment](https://github.com/bennyguo/instant-nsr-pl/issues/60#issuecomment-1515214611) to use view-independent color the model...

Hi! Theoretically, FullyFusedMLP should be faster than VanillaMLP, so we use it for NeRF (although we do not observe significant speedup lol). But for NeuS, computing the analytic normal requires...

Same here. I'm using `torch==1.12.1+cu113` and `xformers=0.0.20+6425fd0.d20230429`. xformers is built from source.

Was the guidance successfully loaded? Also you may directly kill the process and see where it gets stuck.

Hi Joseph, thank you for the contribution! I saw you're using stable-diffusion-guidance instead of stable-diffusion-unified-guidance for the HiFA implementation, where the latter provides useful options like return_rgb_1step_orig and return_rgb_multistep_orig which...

@JunzheJosephZhu Sorry for the inconvenience. stable-diffusion-unified-guidance covers all the functionality of stable-diffusion-guidance and ideally should be used in all systems. Feel free to use it in HiFA system and I'll...

see this issue https://github.com/threestudio-project/threestudio/issues/350