xuyu666

Hello author, I am very happy to be able to reproduce your code. I am very curious about your thoughts on design loss, I want to ask some questions about...

The demo environment is incorrectly configured, and the webpage is not Found.

1

Hello author, I encountered some problems while using your model. I did the following steps: 1. Run start.sh according to the instructions, and then the npm command reported a lot...

如何利用c站的safetensor生成？

1

您好， 想了解下从c站下载的约6GB的safetensor sdxl大模型如何应用到当前版本代码中，直接替换貌似不太合理，例如config.json不太对应、模型文件不对应等。想寻求下您的建议。

Hi First of all, thank you for your paper and contribution！ I have a question about 80 blurring kernels of the test dataset. Problem Description： Because the method of my...

UniGS 代码何时公布

3

作者您好： 很高兴看到您的UniGS论文被cvpr 2024接收，我对您这篇工作比较好奇和感兴趣，特别生成和分割统一表示的部分以及之前您做过的高质量分割，请问啥时候可以开源下代码~

Problems with the use of the official website sam(cut out function)

1

After mask labeling, the individuals that came out using the cut out function were not as expected **Original image:**  **Mask labeling:**  **Cropped image:** 

训练相关疑问

12

作者你好： 我最近跑通了你的代码并用于训练。其中，我将你/data/xuyu_model/OOTDiffusion-train/checkpoints/ootd/ootd_hd/checkpoint-36000/unet_garm 内的config和Diffusion.safetensor 直接替换成了SD1.5对应Unet文件内的，这种做法用于重新训练是否正确。同时，在训练过程中监测训练方面，有什么建议么~

作者您好，现在torch不断更新，很多老的包不兼容，例如trilinear，是否可以出一个torch2.0版本的代码，或者如何在torch2.0上使用模型？