Breaking Spatial Uniformity: Prior-Guided Mamba with Radial Serialization for Lens Flare Removal

2026-05-08

AI Summary

Existing lens flare removal methods typically employ spatially uniform processing, which struggles to simultaneously preserve light sources, eliminate artifacts, and restore background details due to their regionally varying requirements. To address this limitation, this work proposes DeflareMambaV2, a novel framework that integrates flare prior guidance with a radial serialization strategy to enable spatially non-uniform, long-range adaptive restoration within a Mamba-based state space model. The approach leverages a prior network to deliver region-aware guidance and incorporates a curriculum-based pixel-level intensity calibration scheme, substantially enhancing restoration fidelity. Experimental results demonstrate that the proposed method outperforms current state-of-the-art techniques across multiple quantitative metrics while significantly reducing model parameter count.

Abstract

Lens flares, caused by complex optical aberrations, severely degrade image quality especially in nighttime photography. Although recent restoration methods have made remarkable progress, most still rely on spatially uniform processing. They are failing to handle the region-dependent restoration demands of flare scenes, where saturated light sources should be preserved, flare artifacts removed, and background details recovered. To address this challenge, we propose DeflareMambav2, a prior-guided Mamba framework for lens flare removal. Specifically, we introduce a Flare Prior Network (FPN) to estimate flare priors and guide adaptive restoration. Besides, a novel radial serialization strategy breaks spatially homogeneous processing by performing flare-aware targeted sampling, and better supports long-range modeling in State Space Models (SSMs). Based on these priors, the backbone adopts a dual-level adaptive scheme. It explicitly preserves light-source regions to avoid over-processing, and applies curriculum-based restoration to the remaining contaminated areas while calibrating restoration intensity at the pixel level. Extensive experiments demonstrate that DeflareMambav2 achieves state-of-the-art performance with reduced parameter burden. Code is available at https://github.com/BNU-ERC-ITEA/DeflareMambav2.

Problem

• lens flare removal
• spatial uniformity
• region-dependent restoration
• flare artifacts
• light source preservation

Innovation

• prior-guided Mamba
• radial serialization
• flare removal
• spatial non-uniform processing
• State Space Models