Collects shadow images for multiple scenarios and compiled a new dataset of 10,500 shadow images, each with labeled ground-truth mask, for supporting shadow detection in the complex world. The dataset covers a rich variety of scene categories, with diverse shadow sizes, locations, contrasts, and types.
Source: Revisiting Shadow Detection: A New Benchmark Dataset for Complex World
Variants: CUHK-Shadow
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Shadow Detection | SDDNet (MM 2023) (512x512) | SDDNet: Style-guided Dual-layer Disentanglement Network … | 2023-08-17 |
| Shadow Detection | SDDNet (MM 2023) (256x256) | SDDNet: Style-guided Dual-layer Disentanglement Network … | 2023-08-17 |
| Shadow Detection | FSDNet (TIP 2021) (512x512) | Revisiting Shadow Detection: A New … | 2019-11-16 |
| Shadow Detection | FSDNet (TIP 2021) (256x256) | Revisiting Shadow Detection: A New … | 2019-11-16 |
| Shadow Detection | DSC (CVPR 2018, TPAMI 2020) (512x512) | Direction-aware Spatial Context Features for … | 2018-05-12 |
| Shadow Detection | DSC (CVPR 2018, TPAMI 2020) (256x256) | Direction-aware Spatial Context Features for … | 2018-05-12 |
Recent papers with results on this dataset: