-
As the title, can we replace stable diffusion by deepfloyd in prolificdreamer?
-
Hi, thanks for your great repo! I have tried prolific-dreamer and get a nice result.
However, the normal map seems weird:
https://github.com/threestudio-project/threestudio/assets/52091468/b4cc56…
-
Why lora's noise needs to do the following calculation and then calculate the loss with SD's noise;
https://github.com/threestudio-project/threestudio/blob/1de1f157aaf34307bb65c05c6bba9a46a03efdcc/th…
-
May make:
https://dreamfusion3d.github.io/
https://github.com/KU-CVLAB/3DFuse
https://github.com/openai/point-e
as alternative generator engines?
-
Classifier free guidance should be calculated as `noise_pred = uncond + cfg * (cond - uncond)`, but it seems all the guidance modules are implemented as `noise_pred = cond + cfg * (cond - uncond)`. e.…
-
https://github.com/threestudio-project/threestudio/blob/abfe2139cb4728bd1447f4ca85d81fdc7bf414e0/README.md?plain=1#L429
`system.prompt_processor.enable_prompt_debiasing=true` >> `system.prompt_proces…
-
Hello.
I've set up env with docker like this:
```
FROM nvidia/cuda:11.8.0-cudnn8-devel-ubuntu20.04
ENV TORCH_CUDA_ARCH_LIST="8.0"
ENV TCNN_CUDA_ARCHITECTURES=80
ENV CUDA_HOME=/usr/local/cuda…
-
The new work [Magic123](https://arxiv.org/pdf/2306.17843.pdf) seems interesting. Most of the techniques are already implemented here so it might now be much more work.
-
Thanks for the great job for prolificdreamer and DreamFusion! I have a question about the calculation of SDS loss. I found in the paper `g = matmul(weight(t) * (-epshat t - eps), grad(x, params))`.
…
-
Thank you a lot for your nice share!
I try to train magic3d with prompt "ironman, full body".
4 V100 cards and batchsize = 2 for each card
After the coarse stage of magic3d training，the visual …