Low-Light Video Enhancement with An Effective Spatial-Temporal Decomposition Paradigm

2026-02-09

AI Summary

This work proposes VLLVE, a view-aware low-light video enhancement framework designed to address the severe visual degradation caused by extreme underexposure and noise in low-light videos. By decoupling view-invariant (intrinsic appearance) and view-dependent (illumination) components, VLLVE introduces a spatial-temporal decomposition paradigm through a dual-structured encoder-decoder architecture with additive residual blocks. The framework jointly optimizes enhancement and degradation processes by incorporating cross-frame correspondence modeling, scene continuity constraints, and an end-to-end bidirectional learning mechanism. Extensive experiments on standard benchmarks demonstrate that VLLVE achieves superior robustness and visual quality in real-world complex dynamic scenes.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Learning to Search

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

Low-Light Video Enhancement (LLVE) seeks to restore dynamic or static scenes plagued by severe invisibility and noise. In this paper, we present an innovative video decomposition strategy that incorporates view-independent and view-dependent components to enhance the performance of LLVE. The framework is called View-aware Low-light Video Enhancement (VLLVE). We leverage dynamic cross-frame correspondences for the view-independent term (which primarily captures intrinsic appearance) and impose a scene-level continuity constraint on the view-dependent term (which mainly describes the shading condition) to achieve consistent and satisfactory decomposition results. To further ensure consistent decomposition, we introduce a dual-structure enhancement network featuring a cross-frame interaction mechanism. By supervising different frames simultaneously, this network encourages them to exhibit matching decomposition features. This mechanism can seamlessly integrate with encoder-decoder single-frame networks, incurring minimal additional parameter costs. Building upon VLLVE, we propose a more comprehensive decomposition strategy by introducing an additive residual term, resulting in VLLVE++. This residual term can simulate scene-adaptive degradations, which are difficult to model using a decomposition formulation for common scenes, thereby further enhancing the ability to capture the overall content of videos. In addition, VLLVE++ enables bidirectional learning for both enhancement and degradation-aware correspondence refinement (end-to-end manner), effectively increasing reliable correspondences while filtering out incorrect ones. Notably, VLLVE++ demonstrates strong capability in handling challenging cases, such as real-world scenes and videos with high dynamics. Extensive experiments are conducted on widely recognized LLVE benchmarks.

Problem

• Low-Light Video Enhancement
• video decomposition
• noise
• invisibility
• scene degradation

Innovation

• Spatial-Temporal Decomposition
• View-aware Enhancement
• Cross-frame Correspondence
• Additive Residual Modeling
• Bidirectional Learning