WavShadow: Wavelet Based Shadow Segmentation and Removal

TL;DR

WavShadow introduces a wavelet-based shadow segmentation and removal method that leverages MAE priors, Fourier features, and a modified SAM Adapter to improve accuracy and convergence speed in complex scenes.

Contribution

The paper presents a novel wavelet-based approach combining MAE priors, Fourier features, and a modified SAM Adapter for enhanced shadow segmentation and removal.

Findings

Achieves state-of-the-art results on DESOBA dataset

Faster convergence compared to previous methods

Improved shadow removal quality in complex scenarios

Abstract

Shadow removal and segmentation remain challenging tasks in computer vision, particularly in complex real world scenarios. This study presents a novel approach that enhances the ShadowFormer model by incorporating Masked Autoencoder (MAE) priors and Fast Fourier Convolution (FFC) blocks, leading to significantly faster convergence and improved performance. We introduce key innovations: (1) integration of MAE priors trained on Places2 dataset for better context understanding, (2) adoption of Haar wavelet features for enhanced edge detection and multiscale analysis, and (3) implementation of a modified SAM Adapter for robust shadow segmentation. Extensive experiments on the challenging DESOBA dataset demonstrate that our approach achieves state of the art results, with notable improvements in both convergence speed and shadow removal quality.