Single Image Reflection Removal via inter-layer Complementarity

2025-05-19

AI Summary

In single-image reflection removal, conventional dual-stream architectures neglect inter-layer complementarity, limiting separation quality. To address this, we propose a physics-consistent inter-layer complementary modeling framework. Our method retains the dual-stream architecture while introducing two key innovations: (1) a novel residual low-frequency–transmission-layer co-modeling mechanism coupled with high-frequency inverse modulation, enhancing frequency-domain physical interpretability; and (2) channel-wise cross-stream reorganization via inter-layer complementary attention, explicitly capturing structural complementarity between dual-stream features. Extensive experiments demonstrate state-of-the-art performance on benchmark datasets including SIR2 and Real20. Moreover, our approach reduces computational overhead by 23% and model parameters by 18%, achieving superior efficiency and generalization without compromising separation accuracy.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Although dual-stream architectures have achieved remarkable success in single image reflection removal, they fail to fully exploit inter-layer complementarity in their physical modeling and network design, which limits the quality of image separation. To address this fundamental limitation, we propose two targeted improvements to enhance dual-stream architectures: First, we introduce a novel inter-layer complementarity model where low-frequency components extracted from the residual layer interact with the transmission layer through dual-stream architecture to enhance inter-layer complementarity. Meanwhile, high-frequency components from the residual layer provide inverse modulation to both streams, improving the detail quality of the transmission layer. Second, we propose an efficient inter-layer complementarity attention mechanism which first cross-reorganizes dual streams at the channel level to obtain reorganized streams with inter-layer complementary structures, then performs attention computation on the reorganized streams to achieve better inter-layer separation, and finally restores the original stream structure for output. Experimental results demonstrate that our method achieves state-of-the-art separation quality on multiple public datasets while significantly reducing both computational cost and model complexity.

Problem

• Enhance inter-layer complementarity in reflection removal
• Improve detail quality of transmission layer
• Reduce computational cost and model complexity

Innovation

• Inter-layer complementarity model enhances dual-stream architecture
• High-frequency components improve transmission layer detail quality
• Efficient inter-layer complementarity attention mechanism reduces complexity