Learning Adaptive Lighting via Channel-Aware Guidance

TL;DR

This paper introduces LALNet, a multi-task neural network that adaptively handles various lighting challenges by leveraging channel-aware features and attention mechanisms, achieving superior performance with fewer resources.

Contribution

The paper proposes a novel channel-aware multi-task framework, LALNet, which effectively integrates color-separated and mixed features for diverse lighting tasks, advancing the state-of-the-art.

Findings

LALNet outperforms existing methods on multiple benchmarks.

LALNet requires less computational resources.

LALNet effectively handles multiple light-related tasks simultaneously.

Abstract

Learning lighting adaptation is a crucial step in achieving good visual perception and supporting downstream vision tasks. Current research often addresses individual light-related challenges, such as high dynamic range imaging and exposure correction, in isolation. However, we identify shared fundamental properties across these tasks: i) different color channels have different light properties, and ii) the channel differences reflected in the spatial and frequency domains are different. Leveraging these insights, we introduce the channel-aware Learning Adaptive Lighting Network (LALNet), a multi-task framework designed to handle multiple light-related tasks efficiently. Specifically, LALNet incorporates color-separated features that highlight the unique light properties of each color channel, integrated with traditional color-mixed features by Light Guided Attention (LGA). The LGA utilizes color-separated features to guide color-mixed features focusing on channel differences and ensuring visual consistency across all channels. Additionally, LALNet employs dual domain channel modulation for generating color-separated features and a mixed channel modulation and light state space module for producing color-mixed features. Extensive experiments on four representative light-related tasks demonstrate that LALNet significantly outperforms state-of-the-art methods on benchmark tests and requires fewer computational resources. We provide an anonymous online demo at https://xxxxxx2025.github.io/LALNet/.