ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and Removal

TL;DR

ARGAN is a novel deep learning framework that effectively detects and removes shadows in images using an attentive recurrent GAN architecture, improving detail recovery and robustness over existing methods.

Contribution

The paper introduces ARGAN, a new attentive recurrent GAN model with a progressive attention mechanism and semi-supervised training for superior shadow detection and removal.

Findings

Outperforms state-of-the-art methods on four datasets

Effectively detects both simple and complex shadows

Produces more realistic and detailed shadow removal results

Abstract

In this paper we propose an attentive recurrent generative adversarial network (ARGAN) to detect and remove shadows in an image. The generator consists of multiple progressive steps. At each step a shadow attention detector is firstly exploited to generate an attention map which specifies shadow regions in the input image.Given the attention map, a negative residual by a shadow remover encoder will recover a shadow-lighter or even a shadow-free image. A discriminator is designed to classify whether the output image in the last progressive step is real or fake. Moreover, ARGAN is suitable to be trained with a semi-supervised strategy to make full use of sufficient unsupervised data. The experiments on four public datasets have demonstrated that our ARGAN is robust to detect both simple and complex shadows and to produce more realistic shadow removal results. It outperforms the state-of-the-art methods, especially in detail of recovering shadow areas.