Instance Shadow Detection
arXiv:1911.07034
Abstract
Instance shadow detection is a brand new problem, aiming to find shadow instances paired with object instances. To approach it, we first prepare a new dataset called SOBA, named after Shadow-OBject Association, with 3,623 pairs of shadow and object instances in 1,000 photos, each with individual labeled masks. Second, we design LISA, named after Light-guided Instance Shadow-object Association, an end-to-end framework to automatically predict the shadow and object instances, together with the shadow-object associations and light direction. Then, we pair up the predicted shadow and object instances, and match them with the predicted shadow-object associations to generate the final results. In our evaluations, we formulate a new metric named the shadow-object average precision to measure the performance of our results. Further, we conducted various experiments and demonstrate our method's applicability on light direction estimation and photo editing.
arXiv commentAccepted to CVPR 2020
References in corpus (14)
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- Learning to Segment Object Candidates
- Path Aggregation Network for Instance Segmentation
- Direction-aware Spatial Context Features for Shadow Detection
- Direction-aware Spatial Context Features for Shadow Detection and Removal
- Instance-aware Semantic Segmentation via Multi-task Network Cascades
- Learning to Refine Object Segments
- Privacy Protection in Street-View Panoramas using Depth and Multi-View Imagery
- TensorMask: A Foundation for Dense Object Segmentation
- Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal
- Deep Watershed Transform for Instance Segmentation
- MegDet: A Large Mini-Batch Object Detector
- InstanceCut: from Edges to Instances with MultiCut
- Boundary-aware Instance Segmentation