1038lab / SDMatte

SDMatte - SafeTensors Models for Interactive Matting

This repository contains SafeTensors versions of the SDMatte models for interactive image matting, optimized for ComfyUI usage.

About SDMatte

SDMatte: Grafting Diffusion Models for Interactive Matting

SDMatte is a state-of-the-art diffusion-driven interactive matting model that leverages the powerful priors of diffusion models to achieve exceptional performance in extracting fine-grained details, especially in edge regions.

Key Features

• Diffusion-powered : Utilizes diffusion model priors for superior detail extraction
• Interactive matting : Visual prompt-driven interaction for precise control
• Fine-grained details : Excels at capturing complex edge regions and texture details
• Coordinate & opacity awareness : Enhanced spatial and opacity information processing

Available Models

• SDMatte.safetensors - Standard interactive matting model
• SDMatte_plus.safetensors - Enhanced version with improved performance

Credits and Attribution

Original Work

Authors : vivoCameraResearch Team
Original Repository : https://huggingface.co/LongfeiHuang/SDMatte
Official Code : https://github.com/vivoCameraResearch/SDMatte
Paper : SDMatte: Grafting Diffusion Models for Interactive Matting

Abstract

Recent interactive matting methods have shown satisfactory performance in capturing the primary regions of objects, but they fall short in extracting fine-grained details in edge regions. Diffusion models trained on billions of image-text pairs, demonstrate exceptional capability in modeling highly complex data distributions and synthesizing realistic texture details, while exhibiting robust text-driven interaction capabilities, making them an attractive solution for interactive matting.