ComfyUI 图像分区编辑工具包:支持智能消除、物品替换、遮罩编辑、分块处理与严格合成。 | Region-aware image editing toolkit for ComfyUI with smart removal, object replacement, masking,...
ComfyUI-Region-Edit-Toolkit is a set of composable image region processing nodes for ComfyUI. It covers region localization, cropping, mask editing, tiling, color and detail harmonization, and precise pasting of local results. The current version includes 42 RegionEdit... nodes under the Region Edit Toolkit menu. The toolkit provides
such as establishing processing regions from faces, detection boxes, or scan windows; combining automatic masks, hand-painted corrections, and protection masks; planning and merging tiles for large images or multiple discrete targets; and precisely merging local results back to the original image by original coordinates. It also matches local color, low-frequency, high-frequency, and skin micro-texture, and can build and inspect local edit prompts. Optionally, it can use a local Argos model for offline Chinese-to-English translation. Installation requires Python 3.10 or higher, and dependencies must be installed into the Python environment actually used by ComfyUI. The nodes are published to the ComfyUI Registry and can be installed via ComfyUI Manager, Comfy CLI, or Git. Example workflows for smart removal and object replacement are provided. Usage notes emphasize that dimensions and coordinates of images, masks, and crops used in difference analysis and strict composite must correspond. If external edit results change width, height, or crop range, they should first be restored to the original crop size. When dimensions or coordinates are inconsistent, the strict composite node will stop execution and give a clear error. Areas that need to remain unchanged should be explicitly specified via protection masks.
Teams use the toolkit to perform local editing by selecting a region, planning size and cropping, processing the crop with image editing or generation nodes, and then strictly compositing the result back.
Teams use the toolkit to correct masks by hand: automatic masking, then using RegionEditMaskEditorCanvas, PreviewBridge, and MaskEditor to paint or erase, followed by RegionEditMaskEditorApply and feathering or compositing.
Teams use the toolkit to process large images with multiple regions by planning tiles, preparing batches, processing each tile, and merging them back.
Teams use the toolkit for smart removal, as demonstrated in the example workflow that removes target wires while keeping the cat and the rest of the scene at original resolution.
Teams use the toolkit for object replacement, as demonstrated in the example workflow that replaces a coffee cup on a table with a bowl of rice while keeping the spoon and the rest of the scene at original resolution.
✨ Zero-config AI chat assistant. No API key needed — sign up and instantly chat with GPT-5, Claude 4, Gemini 2.5, DeepSeek & 100+ top models. Pay-as-you-go saves you more. Available on Web, iOS, macOS, Android, Linux, Windows. - ChatGPTNextWeb/NextChat
Discover Lorka, a multi-AI tool to combine GPT, Gemini, DeepSeek, and more in a single subscription. Fast, flexible, all-in-one AI.
Download DeskClaw — the easiest way to run OpenClaw AI agent on your desktop. No terminal, no config. Works on Windows & macOS. Free to start.
Official website restored from the pre-incident audit; product details pending editorial verification.