CWNet: Causal Wavelet Network for Low-Light Image Enhancement

2025-07-14

AI Summary

Traditional low-light image enhancement (LLIE) methods often neglect instance-level semantics and frequency-domain characteristics of features. To address this, we propose CausalWaveNet—a novel LLIE framework that pioneers the integration of causal inference into low-light enhancement. Our method decouples causal factors from confounding interferences in brightness variation via global metric learning and local instance-level CLIP-semantic loss. Concurrently, it incorporates wavelet transforms to model and restore luminance and texture details accurately across multiple frequency scales. The end-to-end architecture synergistically optimizes semantic awareness and frequency-domain priors. Extensive experiments on benchmarks including LOL and DARKFACE demonstrate significant improvements over state-of-the-art methods, with average PSNR and SSIM gains of 1.23 dB and 0.021, respectively. Moreover, CausalWaveNet exhibits strong cross-scene generalization capability.

Technology Category

• Machine Learning: Causal Learning
• Computer Vision: Low Level & Physics-based Vision
• Reasoning under Uncertainty: Causality

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
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Traditional Low-Light Image Enhancement (LLIE) methods primarily focus on uniform brightness adjustment, often neglecting instance-level semantic information and the inherent characteristics of different features. To address these limitations, we propose CWNet (Causal Wavelet Network), a novel architecture that leverages wavelet transforms for causal reasoning. Specifically, our approach comprises two key components: 1) Inspired by the concept of intervention in causality, we adopt a causal reasoning perspective to reveal the underlying causal relationships in low-light enhancement. From a global perspective, we employ a metric learning strategy to ensure causal embeddings adhere to causal principles, separating them from non-causal confounding factors while focusing on the invariance of causal factors. At the local level, we introduce an instance-level CLIP semantic loss to precisely maintain causal factor consistency. 2) Based on our causal analysis, we present a wavelet transform-based backbone network that effectively optimizes the recovery of frequency information, ensuring precise enhancement tailored to the specific attributes of wavelet transforms. Extensive experiments demonstrate that CWNet significantly outperforms current state-of-the-art methods across multiple datasets, showcasing its robust performance across diverse scenes. Code is available at https://github.com/bywlzts/CWNet-Causal-Wavelet-Network.

Problem

• Addresses neglect of semantic info in low-light enhancement
• Uses causal reasoning to reveal enhancement relationships
• Optimizes frequency recovery via wavelet transforms

Innovation

• Causal reasoning for low-light image enhancement
• Wavelet transform-based backbone network
• Instance-level CLIP semantic loss