DINOLight: Robust Ambient Light Normalization with Self-supervised Visual Prior Integration

TL;DR

DINOLight introduces a novel ambient light normalization method that leverages self-supervised DINOv2 features to effectively restore images affected by complex lighting and shadows, outperforming existing approaches.

Contribution

The paper proposes a new framework integrating DINOv2 features into ambient light normalization, utilizing adaptive feature fusion and cross-attention for improved image restoration.

Findings

DINOLight outperforms existing methods on Ambient6K dataset.

DINOv2 features significantly enhance lighting normalization.

Achieves competitive results on shadow-removal benchmarks.

Abstract

This paper presents a new ambient light normalization framework, DINOLight, that integrates the self-supervised model DINOv2's image understanding capability into the restoration process as a visual prior. Ambient light normalization aims to restore images degraded by non-uniform shadows and lighting caused by multiple light sources and complex scene geometries. We observe that DINOv2 can reliably extract both semantic and geometric information from a degraded image. Based on this observation, we develop a novel framework to utilize DINOv2 features for lighting normalization. First, we propose an adaptive feature fusion module that combines features from different DINOv2 layers using a point-wise softmax mask. Next, the fused features are integrated into our proposed restoration network in both spatial and frequency domains through an auxiliary cross-attention mechanism. Experiments show that DINOLight achieves superior performance on the Ambient6K dataset, and that DINOv2 features are effective for enhancing ambient light normalization. We also apply our method to shadow-removal benchmark datasets, achieving competitive results compared to methods that use mask priors. Codes will be released upon acceptance.