A+D-Net: Shadow Detection with Adversarial Shadow Attenuation

Single image shadow detection is a very challenging problem because of the limited amount of information available in one image, as well as the scarcity of annotated training data. In this work, we propose a novel adversarial training based framework that yields a high performance shadow detection network (D-Net). D-Net is trained together with an Attenuator network (A-Net) that generates adversarial training examples. A-Net performs shadow attenuation in original training images constrained by a simplified physical shadow model and focused on fooling DNet’s shadow predictions. Hence, it is effectively augmenting the training data for D-Net with hard to predict cases. Experimental results on the most challenging shadow detection benchmark[22] show that our method outperform the state-of-the-art[14] with a 38% error reduction, measured in terms of balanced error rate (BER). Our proposed shadow detector also obtains state-of-the-art results on a cross-dataset task testing on UCF[24] with a 14% error reduction. Furthermore, the proposed method can perform accurate close to real-time shadow detection at a rate of 13 frames per second.