Zhengyi Wang

Thanks, I am looking forward to that.

Thank you for your contribution!

> ```python > latents.data = latents.data.clip(-1, 1) > ``` A very interesting finding! I find it very useful. Thank you very much,

这个是blender自带的功能，直接用blender转换就可以了

Shading in latent space is strange.

https://user-images.githubusercontent.com/40787266/221347907-cdab01e5-40a3-42ba-ab34-db42e14ace61.mp4 My result of hamburger, which is not as good as the one in the paper.

@JDihlmann Thank you for your kind response!

Hi! You can manually call the set_seed() function in the pipeline class. https://github.com/thu-ml/CRM/blob/3e677cb41d4856dc7c46c5dace00d52336f5c614/pipelines.py#L129-#L132

Thank you for your interest in our CRM paper! 1. The decoder cannot directly work at resolution 512x3,072. However, I will upload the model trained on this resolution 512x3,072 if...

Thank you for your interest. I will upload the code for texture rendering :)