https://github.com/AuroBit/ComfyUI-OOTDiffusion
ComfyUI custom node that simply integrates the OOTDiffusion.
https://github.com/AuroBit/ComfyUI-OOTDiffusion
Last synced: 6 months ago
JSON representation
ComfyUI custom node that simply integrates the OOTDiffusion.
- Host: GitHub
- URL: https://github.com/AuroBit/ComfyUI-OOTDiffusion
- Owner: AuroBit
- License: other
- Created: 2024-02-23T11:20:00.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-03-26T02:44:57.000Z (over 1 year ago)
- Last Synced: 2024-04-01T16:24:42.951Z (about 1 year ago)
- Language: Python
- Size: 5.86 MB
- Stars: 214
- Watchers: 7
- Forks: 20
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-comfyui - **ComfyUI OOTDiffusion**
README
# ComfyUI OOTDiffusion
A ComfyUI custom node that simply integrates the [OOTDiffusion](https://github.com/levihsu/OOTDiffusion) functionality.
一个简单接入 OOTDiffusion 的 ComfyUI 节点。
Example workflow: [workflow.json](./assets/workflow.json)
## Instruction 指南
Create environment and download dependencies:
创建环境并下载依赖：
```txt
conda create -n ootd python=3.10
conda activate ootd
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
# Install nvcc compiler for torch cpp extensions
conda install cuda-nvcc -c nvidia
# Clone to custom_nodes
git clone https://github.com/AuroBit/ComfyUI-OOTDiffusion.git custom_nodes/ComfyUI-OOTDiffusion
# Install dependencies
pip install -r custom_nodes/ComfyUI-OOTDiffusion/requirements.txt
```
Simply start ComfyUI and drag the example [workflow.json](./assets/workflow.json).
启动 ComfyUI 并拖入示例 [workflow.json](./assets/workflow.json) 即可。
## Windows 指南
Windows 除了需要参照上面的配置教程，还需要额外配置 MSVC：
首先是 nvcc 安装（确保 VIRTUAL_ENV/bin/nvcc 存在）：
```
conda install cuda-nvcc -c nvidia
```
根据自己系统选择 Windows 10 SDK / Windows 11 SDK.
**注意：不要从 Installer 点击启动 terminal**
进入 `C:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\Build` 目录，
右击进入 terminal，注意要用 CMD 不要 powershell。
启动 native x64 target x64 的编译环境变量（兼容环境下 nvcc 编译会报错）：
```
vcvars64.bat
```
编译完成，成功启动。
## diffusers 版本
`main` 分支锁定 diffusers==0.24
`diffusers-0.26` 分支锁定 diffusers==0.26.x
要切换分支，请使用下面命令：
```
git switch diffusers-0.26
```
并重新安装依赖：
```
pip install --force-reinstall -r custom_nodes/ComfyUI-OOTDiffusion/requirements.txt
```
## FAQ 常见错误
```
fatal error: cuda_runtime.h: No such file or directory compilation terminated.
ninja: build stopped: subcommand failed.
```
解决办法：`conda install cuda-toolkit=12.1 -c nvidia` 并覆写 `CUDA_HOME` `CUDA_PATH` 环境变量
```
subprocess.CalledProcessError: Command '['where', 'cl']' returned non-zero exit status 1.
```
解决办法：仅在 Windows 下出现，根据 [Windows 配置教程](#Windows-指南)
## Node 节点
Load OOTDiffusion Local: 从本地目录加载 OOTDiffusion Pipeline
Load OOTDiffusion from Hub: 从 huggingface 自动下载并加载 OOTDiffusion Pipeline
OOTDiffusion Generate: 生成图像
参数：
cfg: 输出图像和输入衣服的贴合程度
## Example image 示例图片
Half body 半身: [模特](./assets/model_1.png) [衣服](./assets/cloth_1.jpg)
Full body 全身: [模特](./assets/model_fullbody_1.png) [裤子](./assets/cloth_lower_body_1.jpg)
Full body 裙子: [模特](./assets/model_dress_1.png) [裙子](./assets/cloth_dress_1.jpg)
## 更新日志 Release Note
2024-03-14:
添加 `diffusers-0.26` 分支
2024-03-10:
添加 humanparsing onnx 支持
2024-03-04:
添加 Full body 模型
2024-03-01:
添加 Windows 安装教程
2024-02-25:
移除 git lfs 下载教程，提供 Load OOTDiffusion from Hub 节点