Cross-Modal Spherical Aggregation for Weakly Supervised Remote Sensing Shadow Removal

arXiv:2406.17469v1 Announce Type: new
Abstract: Remote sensing shadow removal, which aims to recover contaminated surface information, is tricky since shadows typically display overwhelmingly low illumination intensities. In contrast, the infrared image is robust toward significant light changes, providing visual clues complementary to the visible image. Nevertheless, the existing methods ignore the collaboration between heterogeneous modalities, leading to undesired quality degradation. To fill this gap, we propose a weakly supervised shadow removal network with a spherical feature space, dubbed S2-ShadowNet, to …

abstract aggregation arxiv contrast cs.cv display image information light low modal robust sensing shadow surface type visual