Home
Publications
Photo Gallery
Instance Shadow Detection with a Single-Stage Detector
Tianyu Wang, Xiaowei Hu, Pheng-Ann Heng, Chi-Wing Fu
IEEE TPAMI · 2023
Research summary
Journal extension of CVPR 2021 work with improved single-stage detector for instance shadow detection and shadow-object association.
Abstract
This paper formulates a new problem, instance shadow detection, which aims to detect shadow instance and the associated object instance that cast each shadow in the input image. To approach this task, we first compile a new dataset with the masks for shadow instances, object instances, and shadow-object associations. We then design an evaluation metric for quantitative evaluation of the performance of instance shadow detection. Further, we design a single-stage detector to perform instance shadow detection in an end-to-end manner, where the bidirectional relation learning module and the deformable maskIoU head are proposed in the detector to directly learn the relation between shadow instances and object instances and to improve the accuracy of the predicted masks. Finally, we quantitatively and qualitatively evaluate our method on the benchmark dataset of instance shadow detection and show the applicability of our method on light direction estimation and photo editing.
Key insight
Extends the CVPR 2021 work with bidirectional relation learning and a deformable MaskIoU head, establishing stronger baselines for single-stage instance shadow detection on the SOBA benchmark.
arXiv version linked; publication venue as listed above
Paper & resources
BibTeX
@article{wang2023instance,
title={Instance Shadow Detection with a Single-Stage Detector},
author={Wang, Tianyu and Hu, Xiaowei and Heng, Pheng-Ann and Fu, Chi-Wing},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
volume={45},
number={3},
pages={3259--3273},
year={2023},
publisher={IEEE}
}