|
| 1 | +# ComfyUI_SC_PhotoDoodle |
| 2 | + |
| 3 | +This is a custom node for ComfyUI that implements PhotoDoodle functionality, allowing you to control image generation through simple doodles. |
| 4 | + |
| 5 | +## Original Project Reference |
| 6 | + |
| 7 | +This project is reimplemented based on [PhotoDoodle](https://github.com/vpdonato/PhotoDoodle), providing a more integrated experience for ComfyUI. |
| 8 | + |
| 9 | + |
| 10 | + |
| 11 | +## Example Workflows |
| 12 | + |
| 13 | +Click on the images below to view example workflows: |
| 14 | + |
| 15 | +[](example_workflows/photodoodlev1.json) |
| 16 | + |
| 17 | +[](example_workflows/photodoodle_speedv1.json) |
| 18 | + |
| 19 | +## Node Description |
| 20 | + |
| 21 | +The `PhotoDoodleSamplerAdvanced` node extends the functionality of `customkaspmleAdvance` by adding two parameters: `condition_image` and `use_clone_pe`. The `condition_image` parameter accepts VAE encoded images, and `use_clone_pe` should be set to true for optimal results. |
| 22 | + |
| 23 | +The `PhotoDoodleCrop` node automatically crops any input image to the maximum possible area based on specified width and height. If you don't use this node, you need to ensure that the input image dimensions match the dimensions of the empty latent. |
| 24 | + |
| 25 | +## Installation |
| 26 | + |
| 27 | +1. Clone this repository into the `custom_nodes` directory of ComfyUI |
| 28 | +2. Restart ComfyUI |
| 29 | + |
| 30 | +```bash |
| 31 | +cd custom_nodes |
| 32 | +git clone https://github.com/your-username/ComfyUI_SC_PhotoDoodle.git |
| 33 | +``` |
| 34 | + |
| 35 | +## Usage |
| 36 | + |
| 37 | +1. Load the PhotoDoodle model |
| 38 | +2. Prepare an original image and a doodle image |
| 39 | +3. Use the PhotoDoodleEdit node for editing |
| 40 | +4. Adjust parameters to achieve the desired effect |
| 41 | + |
| 42 | +--- |
| 43 | + |
| 44 | +# ComfyUI_SC_PhotoDoodle |
| 45 | + |
| 46 | +这是一个 ComfyUI 的自定义节点,用于实现 PhotoDoodle 功能,让您可以通过简单的涂鸦来控制图像生成。 |
| 47 | + |
| 48 | +## 原始项目参考 |
| 49 | + |
| 50 | +本项目基于 [PhotoDoodle](https://github.com/vpdonato/PhotoDoodle) 重新实现,为 ComfyUI 提供了更加集成的体验。 |
| 51 | + |
| 52 | + |
| 53 | + |
| 54 | +## 示例工作流 |
| 55 | + |
| 56 | +点击下方图片查看示例工作流: |
| 57 | + |
| 58 | +[](example_workflows/photodoodlev1.json) |
| 59 | + |
| 60 | +[](example_workflows/photodoodle_speedv1.json) |
| 61 | + |
| 62 | +## 节点说明 |
| 63 | + |
| 64 | +`PhotoDoodleSamplerAdvanced` 节点相比于 `customkaspmleAdvance` 增加了 `condition_image` 和 `use_clone_pe` 两个参数,分别传入 VAE 编码后的图片,然后设置 `use_clone_pe` 为 true 以获得最佳效果。 |
| 65 | + |
| 66 | +`PhotoDoodleCrop` 节点可以对任意输入图像,根据设置的宽高,输出最大范围自动裁切后的图片。如果不使用该节点,需要保证传入图片的宽高和创建 `emptylatent` 的宽高保持一致。 |
| 67 | + |
| 68 | +## 安装 |
| 69 | + |
| 70 | +1. 在 ComfyUI 的 `custom_nodes` 目录下克隆此仓库 |
| 71 | +2. 重启 ComfyUI |
| 72 | + |
| 73 | +```bash |
| 74 | +cd custom_nodes |
| 75 | +git clone https://github.com/your-username/ComfyUI_SC_PhotoDoodle.git |
| 76 | +``` |
| 77 | + |
| 78 | +## 使用方法 |
| 79 | + |
| 80 | +1. 加载 PhotoDoodle 模型 |
| 81 | +2. 准备一张原始图像和一张涂鸦图像 |
| 82 | +3. 使用 PhotoDoodleEdit 节点进行编辑 |
| 83 | +4. 调整参数以获得理想效果 |
0 commit comments