ReflexSplit: Single Image Reflection Separation via Layer Fusion-Separation

TL;DR

ReflexSplit introduces a dual-stream framework with innovative fusion and separation modules, achieving state-of-the-art single image reflection separation with improved stability and generalization.

Contribution

It proposes a novel layered fusion-separation approach with cross-scale fusion, differential attention, and curriculum training for enhanced reflection layer disentanglement.

Findings

Achieves state-of-the-art results on synthetic and real-world benchmarks.

Demonstrates superior perceptual quality and robustness.

Outperforms existing methods in reflection separation tasks.

Abstract

Single Image Reflection Separation (SIRS) disentangles mixed images into transmission and reflection layers. Existing methods suffer from transmission-reflection confusion under nonlinear mixing, particularly in deep decoder layers, due to implicit fusion mechanisms and inadequate multi-scale coordination. We propose ReflexSplit, a dual-stream framework with three key innovations. (1) Cross-scale Gated Fusion (CrGF) adaptively aggregates semantic priors, texture details, and decoder context across hierarchical depths, stabilizing gradient flow and maintaining feature consistency. (2) Layer Fusion-Separation Blocks (LFSB) alternate between fusion for shared structure extraction and differential separation for layer-specific disentanglement. Inspired by Differential Transformer, we extend attention cancellation to dual-stream separation via cross-stream subtraction. (3) Curriculum training progressively strengthens differential separation through depth-dependent initialization and epoch-wise warmup. Extensive experiments on synthetic and real-world benchmarks demonstrate state-of-the-art performance with superior perceptual quality and robust generalization. Our code is available at https://github.com/wuw2135/ReflexSplit.