Single-image reflection removal via self-supervised diffusion models
This addresses the problem of image quality degradation due to reflections for computer vision applications, but it is incremental as it builds on existing diffusion models and cycle-consistency techniques.
The paper tackles the problem of removing reflections from single images captured through transparent surfaces, which suffers from a shortage of paired real-world training data, by proposing a hybrid approach combining cycle-consistency with denoising diffusion probabilistic models (DDPM) to effectively remove reflections without requiring paired data, achieving superior performance compared to state-of-the-art methods on datasets like SIR^2, FRR, and a new MRR dataset.
Reflections often degrade the visual quality of images captured through transparent surfaces, and reflection removal methods suffers from the shortage of paired real-world samples.This paper proposes a hybrid approach that combines cycle-consistency with denoising diffusion probabilistic models (DDPM) to effectively remove reflections from single images without requiring paired training data. The method introduces a Reflective Removal Network (RRN) that leverages DDPMs to model the decomposition process and recover the transmission image, and a Reflective Synthesis Network (RSN) that re-synthesizes the input image using the separated components through a nonlinear attention-based mechanism. Experimental results demonstrate the effectiveness of the proposed method on the SIR$^2$, Flash-Based Reflection Removal (FRR) Dataset, and a newly introduced Museum Reflection Removal (MRR) dataset, showing superior performance compared to state-of-the-art methods.