Semi-LAR: Semi-supervised Contrastive Learning with Linear Attention for Removal of Nighttime Flares

2026-05-18

AI Summary

This work addresses the challenges in nighttime lens flare removal—namely, the large spatial extent of flares, their entanglement with scene structure, and the reliance on abundant paired training data—by proposing a semi-supervised removal framework. The method constructs an adaptive pseudo-label bank and refines pseudo-labels through no-reference image quality assessment, momentum updating, and an invalid-label filtering mechanism. Furthermore, it introduces flare-aware patch-level contrastive learning that explicitly leverages flare-contaminated images as negative samples to enhance feature discriminability. Experiments demonstrate that the proposed framework is model-agnostic and achieves significant improvements in both performance and robustness across multiple benchmarks.

Abstract

Lens flare removal is challenging due to the large spatial extent of flare artifacts and their entanglement with scene structures, while existing methods heavily rely on large-scale paired data. We propose a semi-supervised flare removal framework that enables stable learning from unlabeled images by jointly addressing pseudo-label reliability and representation discrimination. We propose an adaptive pseudo-label repository that progressively refines pseudo supervision through no-reference quality assessment, momentum-based updates, and invalid label filtering, effectively mitigating error accumulation. Moreover, we propose a flare-aware contrastive loss that explicitly treats flare-contaminated inputs as negatives and performs patch-level contrastive learning, encouraging representations that are discriminative against flare patterns while remaining consistent with reliable pseudo targets. Extensive experiments on multiple flare benchmarks demonstrate that the proposed framework is model-agnostic and consistently improves performance and robustness.

Problem

• lens flare removal
• semi-supervised learning
• nighttime flares
• unpaired data
• flare artifacts

Innovation

• semi-supervised learning
• contrastive learning
• lens flare removal
• pseudo-label refinement
• linear attention