Generalized Lightness Adaptation with Channel Selective Normalization

TL;DR

This paper introduces Channel Selective Normalization (CSNorm), a novel normalization technique that improves the generalization of lightness adaptation in image processing by selectively normalizing lightness-relevant channels, enabling models to perform well across unknown lightness conditions.

Contribution

The paper proposes CSNorm, a channel filtering normalization method, and an alternating training strategy to enhance lightness adaptation generalization with training on a single lightness condition.

Findings

CSNorm improves generalization to unknown lightness conditions.

Models with CSNorm outperform existing methods on benchmark datasets.

The approach requires training on only one lightness condition.

Abstract

Lightness adaptation is vital to the success of image processing to avoid unexpected visual deterioration, which covers multiple aspects, e.g., low-light image enhancement, image retouching, and inverse tone mapping. Existing methods typically work well on their trained lightness conditions but perform poorly in unknown ones due to their limited generalization ability. To address this limitation, we propose a novel generalized lightness adaptation algorithm that extends conventional normalization techniques through a channel filtering design, dubbed Channel Selective Normalization (CSNorm). The proposed CSNorm purposely normalizes the statistics of lightness-relevant channels and keeps other channels unchanged, so as to improve feature generalization and discrimination. To optimize CSNorm, we propose an alternating training strategy that effectively identifies lightness-relevant channels. The model equipped with our CSNorm only needs to be trained on one lightness condition and can be well generalized to unknown lightness conditions. Experimental results on multiple benchmark datasets demonstrate the effectiveness of CSNorm in enhancing the generalization ability for the existing lightness adaptation methods. Code is available at https://github.com/mdyao/CSNorm.