Self-supervised Automatic Matting

2026-07-11

AI Summary

This work addresses the challenge of high annotation costs for precise alpha matte labels, which limits the scale and generalization of deep image matting models. To overcome this, we propose SSMatte—the first fully self-supervised matting framework that requires no human-labeled data—employing a two-stage strategy of semantic anchoring followed by detail refinement. Our method leverages frozen self-supervised ViT features and introduces a generalized Rayleigh quotient–based semantic anchoring loss alongside a fixed-point optimization mechanism enforcing alpha-RGB consistency. Experiments demonstrate that SSMatte achieves performance on par with fully supervised models on portrait matting benchmarks, significantly outperforms weakly supervised approaches, and exhibits strong scalability with increasing data volume as well as robust cross-domain generalization.

Abstract

High-quality alpha mattes are notoriously expensive to annotate, creating a fundamental data bottleneck for deep image matting. While prior work attempts to reduce annotation cost using coarser labels like trimaps or masks, they remain reliant on costly per-pixel supervision, limiting scalability and generalization. In this work, we push the boundary further and ask: can we train an automatic matting model using only RGB images, with no manual annotation at all? We answer this by presenting SSMatte, a self-supervised framework that for the first time achieves performance on par with fully-supervised automatic matting. Our key insight is to decompose the problem into semantic anchoring and detail matting. SSMatte first generates a semantic matting prompt from frozen self-supervised ViT features by propagating class-token seeds via a novel, training-efficient semantic anchoring loss based on a generalized Rayleigh quotient. This prompt then anchors a detail matting network, which is optimized via a fixed-point-based loss that enforces alpha-RGB consistency. Extensive experiments show SSMatte outperforms prior weakly-supervised methods, matches the performance of fully-supervised models on portrait benchmarks, and demonstrates favorable scaling and generalization behaviors with additional data. Our work pushes automatic matting to an fresh, fully annotation-free paradigm. Code will be available.

Problem

• image matting
• self-supervised learning
• annotation-free
• alpha matte
• data bottleneck

Innovation

• self-supervised learning
• image matting
• semantic anchoring
• alpha matte
• annotation-free