PhaSR: Generalized Image Shadow Removal with Physically Aligned Priors

TL;DR

PhaSR introduces a novel shadow removal method that uses physically aligned priors and dual-level alignment techniques to improve robustness across diverse lighting conditions.

Contribution

It proposes Physically Aligned Normalization and Geometric-Semantic Rectification Attention for better shadow removal under complex illumination.

Findings

Achieves competitive shadow removal performance.

Handles multi-source ambient lighting effectively.

Lower complexity compared to traditional methods.

Abstract

Shadow removal under diverse lighting conditions requires disentangling illumination from intrinsic reflectance, a challenge compounded when physical priors are not properly aligned. We propose PhaSR (Physically Aligned Shadow Removal), addressing this through dual-level prior alignment to enable robust performance from single-light shadows to multi-source ambient lighting. First, Physically Aligned Normalization (PAN) performs closed-form illumination correction via Gray-world normalization, log-domain Retinex decomposition, and dynamic range recombination, suppressing chromatic bias. Second, Geometric-Semantic Rectification Attention (GSRA) extends differential attention to cross-modal alignment, harmonizing depth-derived geometry with DINO-v2 semantic embeddings to resolve modal conflicts under varying illumination. Experiments show competitive performance in shadow removal with lower complexity and generalization to ambient lighting where traditional methods fail under multi-source illumination. Our source code is available at https://github.com/ming053l/PhaSR.