chflame163/ComfyUI_LayerStyle

A set of nodes for ComfyUI that can composite layer and mask to achieve Photoshop like functionality.

What it solves

ComfyUI Layer Style provides a set of nodes that bring Photoshop-like compositing and masking capabilities directly into ComfyUI. It aims to centralize the image editing workflow, reducing the need for users to switch between ComfyUI and external image editing software.

How it works

The project implements various nodes for layer and mask manipulation. It integrates several advanced AI models for automated masking and prompt generation, including SAM2, Florence-2, and Llama Vision, while providing traditional image processing tools like Gaussian blur, color inversion, and LUT application.

Who it’s for

Digital artists and AI image generators using ComfyUI who want professional-grade layer management, precise masking, and image compositing tools within a single node-based environment.

Highlights

  • Photoshop-like Compositing: Nodes for blending, drop shadows, and layer management.
  • Advanced Masking: Integration of SAM2, BiRefNet, and Segment Anything for high-precision masks.
  • ** uma Prompt Generation**: Local inference nodes for Llama 3.2 Vision and JoyCaption2 to generate image prompts.
  • Utility Tools: Features for image batching, rounded rectangle masks, and color tone extraction.

Related

  • Project
  • Project
  • Project
  • Project
  • Project