BiEvLight: Bi-level Learning of Task-Aware Event Refinement for Low-Light Image Enhancement

2026-03-05

AI Summary

This work addresses the challenge of modality fusion noise coupling in low-light image enhancement with event cameras, which arises from background activity noise in events and the low signal-to-noise ratio of intensity images. To this end, we propose a task-aware bilevel optimization framework that jointly optimizes event denoising and image enhancement. By modeling the gradient correlation between images and events, we introduce a gradient-guided denoising prior that dynamically adapts event denoising to the requirements of the enhancement task. Our approach is the first to formulate event denoising as a bilevel optimization problem constrained by the downstream enhancement objective, thereby overcoming the limitations of conventional static preprocessing pipelines. Experiments on the SDE real-world noisy dataset demonstrate significant improvements over existing methods, with gains of 1.30 dB in PSNR, 2.03 dB in PSNR*, and 0.047 in SSIM.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Privacy-enhancing technologies

Abstract

Event cameras, with their high dynamic range, show great promise for Low-light Image Enhancement (LLIE). Existing works primarily focus on designing effective modal fusion strategies. However, a key challenge is the dual degradation from intrinsic background activity (BA) noise in events and low signal-to-noise ratio (SNR) in images, which causes severe noise coupling during modal fusion, creating a critical performance bottleneck. We therefore posit that precise event denoising is the prerequisite to unlocking the full potential of event-based fusion. To this end, we propose BiEvLight, a hierarchical and task-aware framework that collaboratively optimizes enhancement and denoising by exploiting their intrinsic interdependence. Specifically, BiEvLight exploits the strong gradient correlation between images and events to build a gradient-guided event denoising prior that alleviates insufficient denoising in heavily noisy regions. Moreover, instead of treating event denoising as a static pre-processing stage-which inevitably incurs a trade-off between over- and under-denoising and cannot adapt to the requirements of a specific enhancement objective-we recast it as a bilevel optimization problem constrained by the enhancement task. Through cross-task interaction, the upper-level denoising problem learns event representations tailored to the lower-level enhancement objective, thereby substantially improving overall enhancement quality. Extensive experiments on the Real-world noise Dataset SDE demonstrate that our method significantly outperforms state-of-the-art (SOTA) approaches, with average improvements of 1.30dB in PSNR, 2.03dB in PSNR* and 0.047 in SSIM, respectively. The code will be publicly available at https://github.com/iijjlk/BiEvlight.

Problem

• event camera
• low-light image enhancement
• background activity noise
• signal-to-noise ratio
• noise coupling

Innovation

• bilevel optimization
• task-aware denoising
• event camera
• low-light image enhancement
• gradient-guided prior