commit BenUltra and LoadBenModel nodes

chflame163 · Nov 17, 2024 · 1ba0eaf · 1ba0eaf
1 parent 7ad0e97
commit 1ba0eaf
Show file tree

Hide file tree

Showing 8 changed files with 1,157 additions and 1 deletion.
diff --git a/README.MD b/README.MD
@@ -150,6 +150,12 @@ Please try downgrading the ```protobuf``` dependency package to 3.20.3, or set e
 
 <font size="4">**If the dependency package error after updating,  please double clicking ```repair_dependency.bat``` (for Official ComfyUI Protable) or  ```repair_dependency_aki.bat``` (for ComfyUI-aki-v1.x) in the plugin folder to reinstall the dependency packages. </font><br />    
 
+
+* Commit [BenUltra](#BenUltra) and [LoadBenModel](#LoadBenModel) nodes. These two nodes are the implementation of [PramaLLC/BEN](https://huggingface.co/PramaLLC/BEN) project in ComfyUI.   
+Download the model files from [huggingface](https://huggingface.co/PramaLLC/BEN/tree/main) or [BaiduNetdisk](https://pan.baidu.com/s/17mdBxfBl_R97mtNHuiHsxQ?pwd=2jn3) and copy to ```ComfyUI/models/BEN``` folder.
+* Merge the PR submitted by [jimlee2048](https://github.com/jimlee2048), add the LoadBiRefNetModelV2 node, and support loading RMBG 2.0 models.       
+Download the model files from [huggingface](https://huggingface.co/briaai/RMBG-2.0/tree/main) or [BaiduNetdisk](https://pan.baidu.com/s/1viIXlZnpTYTKkm2F-QMj_w?pwd=axr9) and copy to ```ComfyUI/models/BiRefNet/RMBG-2.0``` folder.
+
 * Florence2 nodes support base-PromptGen-v2.0 and large-PromptGen-v2.0, Download ```base-PromptGen-v2.0``` and ```large-PromptGen-v2.0``` two folder from [huggingface](https://huggingface.co/chflame163/ComfyUI_LayerStyle/tree/main/ComfyUI/models/florence2) or [BaiduNetdisk](https://pan.baidu.com/s/1BVvXt3N7zrBnToyF-GrC_A?pwd=xm0x) and copy to ```ComfyUI/models/florence2``` folder.
 * [SAM2Ultra](#SAM2Ultra) and ObjectDetector nodes support image batch.
 * [SAM2Ultra](#SAM2Ultra) and [SAM2VideoUltra](#SAM2VideoUltra) nodes add support for SAM2.1 model, including [kijai](https://github.com/kijai)'s FP16 model. Download model files from [BaiduNetdisk](https://pan.baidu.com/s/1xaQYBA6ktxvAxm310HXweQ?pwd=auki) or [huggingface.co/Kijai/sam2-safetensors](https://huggingface.co/Kijai/sam2-safetensors/tree/main) and copy to ```ComfyUI/models/sam2``` folder.
@@ -2341,6 +2347,38 @@ On the basis of RemBgUltra, the following changes have been made:
 * device: Set whether the VitMatte to use cuda.
 * max_megapixels: Set the maximum size for VitMate operations.
 
+
+
+### <a id="table1">BenUltra</a>
+It is the implementation of [PramaLLC/BEN](https://huggingface.co/PramaLLC/BEN)  project in ComfyUI. Thank you to the original author.
+
+Download the model files from [huggingface](https://huggingface.co/PramaLLC/BEN/tree/main) or [BaiduNetdisk](https://pan.baidu.com/s/17mdBxfBl_R97mtNHuiHsxQ?pwd=2jn3) and copy to ```ComfyUI/models/BEN``` folder.
+
+![image](image/ben_ultra_example.jpg)
+
+Node Options：
+![image](image/ben_ultra_node.jpg)
+* ben_model: Ben model input.
+* image: Image input.
+* detail_method: Edge processing methods. provides VITMatte, VITMatte(local), PyMatting, GuidedFilter. If the model has been downloaded after the first use of VITMatte, you can use VITMatte (local) afterwards.
+* detail_erode: Mask the erosion range inward from the edge. the larger the value, the larger the range of inward repair.
+* detail_dilate: The edge of the mask expands outward. the larger the value, the wider the range of outward repair.
+* black_point: Edge black sampling threshold.
+* white_point: Edge white sampling threshold.
+* process_detail: Set to false here will skip edge processing to save runtime.
+* max_megapixels: Set the maximum size for VitMate operations.
+
+### <a id="table1">LoadBenModel</a>
+Load the BEN model. 
+
+
+Node Options:  
+![image](image/load_ben_model_node.jpg)  
+
+* model: Select the model. Currently, only the Ben_Sase model is available for selection.
+
+
+
 ### <a id="table1">BiRefNetUltra</a>
 
 Using the BiRefNet model to remove background has better recognition ability and ultra-high edge details.
@@ -2392,6 +2430,18 @@ Node Options:
 
 * model: Select the model. List the files in the  ```CoomfyUI/models/BiRefNet/pth```  folder for selection.
 
+
+### <a id="table1">LoadBiRefNetModelV2</a>
+This node is a PR submitted by [jimlee2048](https://github.com/jimlee2048) and supports loading RMBG-2.0 models.
+
+Download model files from [huggingface](https://huggingface.co/briaai/RMBG-2.0/tree/main) or [百度网盘](https://pan.baidu.com/s/1viIXlZnpTYTKkm2F-QMj_w?pwd=axr9) and copy to ```ComfyUI/models/BiRefNet/RMBG-2.0``` folder.
+
+Node Options:  
+![image](image/load_birefnet_model_v2_node.jpg)  
+
+* model: Select the model. There are two options, ```BiRefNet-General``` and ```RMBG-2.0```. 
+
+
 ### <a id="table1">TransparentBackgroundUltra</a>
 
 Using the transparent-background model to remove background has better recognition ability and speed, while also having ultra-high edge details.

diff --git a/README_CN.MD b/README_CN.MD
@@ -127,6 +127,10 @@ If this call came from a _pb2.py file, your generated code is out of date and mu
 ## 更新说明
 <font size="4">**如果本插件更新后出现依赖包错误，请双击运行插件目录下的```install_requirements.bat```(官方便携包)，或 ```install_requirements_aki.bat```(秋叶整合包) 重新安装依赖包。
 
+* 添加[BenUltra](#BenUltra) 和 [LoadBenModel](#LoadBenModel)节点。这两个节点是[PramaLLC/BEN](https://huggingface.co/PramaLLC/BEN) 项目在ComfyUI中的实现。    
+从 [huggingface](https://huggingface.co/PramaLLC/BEN/tree/main) 或 [百度网盘](https://pan.baidu.com/s/17mdBxfBl_R97mtNHuiHsxQ?pwd=2jn3)下载模型文件并复制到 ```ComfyUI/models/BEN```文件夹。
+* 合并[jimlee2048](https://github.com/jimlee2048)提交的PR, 添加[LoadBiRefNetModelV2](#LoadBiRefNetModelV2) 节点，支持加载RMBG 2.0模型。    
+从 [huggingface](https://huggingface.co/briaai/RMBG-2.0/tree/main) 或 [百度网盘](https://pan.baidu.com/s/1viIXlZnpTYTKkm2F-QMj_w?pwd=axr9) 下载全部文件并复制到```ComfyUI/models/BiRefNet/RMBG-2.0```文件夹。
 * Florence2 节点更新支持 base-PromptGen-v2.0 和 large-PromptGen-v2.0 模型。从 [huggingface](https://huggingface.co/chflame163/ComfyUI_LayerStyle/tree/main/ComfyUI/models/florence2) 或 [BaiduNetdisk](https://pan.baidu.com/s/1BVvXt3N7zrBnToyF-GrC_A?pwd=xm0x) 下载  ```base-PromptGen-v2.0``` 和 ```large-PromptGen-v2.0``` 两个文件夹并复制到 ```ComfyUI/models/florence2```。
 * [SAM2Ultra](#SAM2Ultra) 及 ObjectDetector 节点支持图像批次。
 * [SAM2Ultra](#SAM2Ultra) 及 [SAM2VideoUltra](#SAM2VideoUltra) 节点增加支持SAM2.1模型，包括[kijai](https://github.com/kijai)量化版fp16模型。请从请从[百度网盘](https://pan.baidu.com/s/1xaQYBA6ktxvAxm310HXweQ?pwd=auki) 或者 [huggingface.co/Kijai/sam2-safetensors](https://huggingface.co/Kijai/sam2-safetensors/tree/main)下载模型文件并复制到```ComfyUI/models/sam2```文件夹。 
@@ -2097,6 +2101,34 @@ RemBgUltra的V2升级版，增加了VITMatte边缘处理方法。
 * device: 设置是否使用cuda。
 * max_megapixels: 设置vitmatte运算的最大尺寸。
 
+### <a id="table1">BenUltra</a>
+是[PramaLLC/BEN](https://huggingface.co/PramaLLC/BEN) 项目在ComfyUI中的实现。感谢原作者。    
+从 [huggingface](https://huggingface.co/PramaLLC/BEN/tree/main) 或 [百度网盘](https://pan.baidu.com/s/17mdBxfBl_R97mtNHuiHsxQ?pwd=2jn3)下载模型文件并复制到 ```ComfyUI/models/BEN```文件夹。
+
+![image](image/ben_ultra_example.jpg)
+
+节点选项说明：
+![image](image/ben_ultra_node.jpg)
+* ben_model: Ben模型输入。
+* image: 图片输入。
+* detail_method: 边缘处理方法。提供了VITMatte, VITMatte(local), PyMatting, GuidedFilter。如果首次使用VITMatte后模型已经下载，之后可以使用VITMatte(local)。
+* detail_erode: 遮罩边缘向内侵蚀范围。数值越大，向内修复的范围越大。
+* detail_dilate: 遮罩边缘向外扩张范围。数值越大，向外修复的范围越大。
+* black_point: 边缘黑色采样阈值。
+* white_point: 边缘白色采样阈值。
+* process_detail: 此处设为False将跳过边缘处理以节省运行时间。
+* max_megapixels: 设置vitmatte运算的最大尺寸。
+
+### <a id="table1">LoadBenModel</a>
+加载BEN模型。 
+
+
+节点选项说明:  
+![image](image/load_ben_model_node.jpg)  
+
+* model: 选择模型。目前仅有Ben_Base 模型可选。
+
+
 ### <a id="table1">BiRefNetUltra</a>
 使用BiRefNet模型去除背景，有更好的识别能力，同时具有超高的边缘细节。
 本节点模型部分的代码来自vipery的[ComfyUI-BiRefNet](https://github.com/viperyl/ComfyUI-BiRefNet)，感谢原作者。
@@ -2147,6 +2179,17 @@ RemBgUltra的V2升级版，增加了VITMatte边缘处理方法。
 
 * model: 选择模型。列出 ```CoomfyUI/models/BiRefNet/pth``` 文件夹下的文件供选择。
 
+### <a id="table1">LoadBiRefNetModelV2</a>
+本节点是[jimlee2048](https://github.com/jimlee2048)提交的PR，支持加载RMBG-2.0模型。    
+从 [huggingface](https://huggingface.co/briaai/RMBG-2.0/tree/main) 或 [百度网盘](https://pan.baidu.com/s/1viIXlZnpTYTKkm2F-QMj_w?pwd=axr9) 下载全部文件并复制到```ComfyUI/models/BiRefNet/RMBG-2.0```文件夹。
+
+节点选项说明:  
+![image](image/load_birefnet_model_v2_node.jpg)  
+
+* model: 选择模型。有两个选项： ```BiRefNet-General``` 和 ```RMBG-2.0```。
+
+
+
 ### <a id="table1">TransparentBackgroundUltra</a>
 使用transparent-background模型去除背景，有更好的识别能力和识别速度，同时具有超高的边缘细节。
 

diff --git a/image/ben_ultra_example.jpg b/image/ben_ultra_example.jpg
diff --git a/image/ben_ultra_node.jpg b/image/ben_ultra_node.jpg
diff --git a/image/load_ben_model_node.jpg b/image/load_ben_model_node.jpg
diff --git a/image/load_birefnet_model_v2_node.jpg b/image/load_birefnet_model_v2_node.jpg