RIDE: Retinex-Informed Decoupling for Exposing Concealed Objects

Chunming He, Rihan Zhang, Dingming Zhang, Chengyu Fang, Longxiang Tang, Jingjia Feng, Fengyang Xiao, Sina Farsiu

arXiv:2605.15450 · 2026-05-18 공개 · arXiv · PDF

contrastive-loss concealed-object-segmentation illumination-reflectance discriminability-gap task-driven-decomposition camouflage-breaking-loss image-decomposition vision-decomposition

Abstract

Concealed Object Segmentation (COS) encompasses a family of dense-prediction tasks, including camouflaged object detection, polyp segmentation, transparent object detection, and industrial defect inspection, where targets are visually entangled with their surroundings through different physical mechanisms. Existing methods either operate directly on RGB images or employ \emph{heterogeneous} decompositions (\eg, Fourier, wavelet) that redistribute spatial evidence across scale/frequency coefficients, making pixel-aligned cues less direct. We introduce a fundamentally different perspective: \textbf{homogeneous image decomposition} via Retinex theory, which factorizes an image into illumination and reflectance components within the \emph{same} spatial domain. Our key insight is that visual entanglement enforces appearance matching in the composite space, but this does \emph{not} necessitate simultaneous matching in both component spaces, a phenomenon we formalize as the \textbf{Discriminability Gap Theorem}. Crucially, we show that across diverse COS sub-tasks, the underlying physical processes systematically anti-correlate illumination and reflectance differences, yielding theoretical guarantees that Retinex decomposition preserves or strictly improves total foreground--background discriminability across the full physical regime, with anti-correlation maximizing the gain. Building on this, we propose \textbf{RIDE} comprising: (i) a Task-Driven Retinex Decomposition module that learns segmentation-optimal factorizations end-to-end; (ii) a Discriminability Gap Attention mechanism that adaptively exploits where decomposition helps; and (iii) a Camouflage-Breaking Contrastive loss operating in reflectance feature space.

한국어 요약

📋 한 줄 요약

**[Computer Vision / Segmentation]** Retinex 이론 기반 동질적 이미지 분해로 위장·은폐 객체 분할 성능을 향상시키는 RIDE.

🎯 핵심 기여도

💡 핵심 아이디어

기존 방법은 RGB를 직접 사용하거나 푸리에·웨이블릿 같은 **이질적** 분해를 사용해 픽셀 정렬 단서가 약화된다. Retinex는 같은 공간 도메인 내에서 조도와 반사도로 분해하는 **동질적** 분해다 — 합성 공간에서는 위장이 외관을 일치시키지만 두 성분에서 동시에 일치할 필요는 없으므로 분해를 통해 판별성이 회복된다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: COS 전 영역을 통일적으로 다루는 이론적 토대와 실용 아키텍처를 동시에 제공한다. **한계**: 이론적 보장이 반상관 가정에 의존하며, 매우 강한 조명 불변 위장에서는 이득이 제한될 수 있다.

🚀 실용적 활용