DFDNet: Dynamic Frequency-Guided De-Flare Network

2025-07-23

AI Summary

In low-light photography, large-scale halos induced by intense light sources severely degrade image quality and impair downstream vision tasks. Existing methods struggle to simultaneously suppress extensive halos and restore structural details near the light source. To address this, we propose the Dynamic Frequency-Guided Halo Removal Network (DFG-Net). Our approach is the first to exploit the pronounced spectral divergence between halos and authentic scene content in the frequency domain: a Dynamic Global Frequency Guidance module enables effective decoupling, while a Local Detail Guidance module preserves fine-grained structures. We further integrate dynamic frequency feature optimization, contrastive learning–driven local feature alignment, and joint frequency-spatial domain modeling. Extensive experiments demonstrate that DFG-Net achieves significant improvements over state-of-the-art methods across multiple benchmarks. The source code is publicly available.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Local Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Strong light sources in nighttime photography frequently produce flares in images, significantly degrading visual quality and impacting the performance of downstream tasks. While some progress has been made, existing methods continue to struggle with removing large-scale flare artifacts and repairing structural damage in regions near the light source. We observe that these challenging flare artifacts exhibit more significant discrepancies from the reference images in the frequency domain compared to the spatial domain. Therefore, this paper presents a novel dynamic frequency-guided deflare network (DFDNet) that decouples content information from flare artifacts in the frequency domain, effectively removing large-scale flare artifacts. Specifically, DFDNet consists mainly of a global dynamic frequency-domain guidance (GDFG) module and a local detail guidance module (LDGM). The GDFG module guides the network to perceive the frequency characteristics of flare artifacts by dynamically optimizing global frequency domain features, effectively separating flare information from content information. Additionally, we design an LDGM via a contrastive learning strategy that aligns the local features of the light source with the reference image, reduces local detail damage from flare removal, and improves fine-grained image restoration. The experimental results demonstrate that the proposed method outperforms existing state-of-the-art methods in terms of performance. The code is available at href{https://github.com/AXNing/DFDNet}{https://github.com/AXNing/DFDNet}.

Problem

• Removes large-scale flare artifacts in nighttime images
• Repairs structural damage near light sources
• Decouples flare artifacts from content in frequency domain

Innovation

• Dynamic frequency-guided flare removal network
• Global dynamic frequency-domain guidance module
• Local detail guidance via contrastive learning