HyperAIHyperAI
HomeConsoleDocsNewsPapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAIHyperAI

Command Palette

Search for a command to run...

Shadow Detection On Sbu

Metrics

BER

Results

Performance results of various models on this benchmark

Paper TitleRepository
ECA (MM 2021) (512x512)7.52Mitigating Intensity Bias in Shadow Detection via Feature Decomposition and Reweighting-
FSDNet (TIP 2021) (256x256)7.16Revisiting Shadow Detection: A New Benchmark Dataset for Complex World
ECA (MM 2021) (256x256)7.08Mitigating Intensity Bias in Shadow Detection via Feature Decomposition and Reweighting-
FSDNet (TIP 2021) (512x512)6.8Revisiting Shadow Detection: A New Benchmark Dataset for Complex World
DSC (CVPR 2018, TPAMI 2020) (256x256)6.79Direction-aware Spatial Context Features for Shadow Detection and Removal
DSC (CVPR 2018, TPAMI 2020) (512x512)6.34Fast Shadow Detection from a Single Image Using a Patched Convolutional Neural Network
MTMT-Net (CVPR 2020) (256x256)6.32A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection-
BDRAR (ECCV 2018) (256x256)6.27Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection-
MTMT-Net (CVPR 2020) (512x512)5.79A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection-
FDRNet (ICCV 2021) (256x256)5.64Mitigating Intensity Bias in Shadow Detection via Feature Decomposition and Reweighting-
BDRAR (ECCV 2018) (512x512)5.62Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection-
SDDNet (MM 2023) (256x256)5.39SDDNet: Style-guided Dual-layer Disentanglement Network for Shadow Detection
FDRNet (ICCV 2021) (512x512)5.39Mitigating Intensity Bias in Shadow Detection via Feature Decomposition and Reweighting-
DSDNet (CVPR 2019) (256x256)5.37Distraction-Aware Shadow Detection-
DSDNet (CVPR 2019) (512x512)5.04Distraction-Aware Shadow Detection-
SDDNet (MM 2023) (512x512)4.86SDDNet: Style-guided Dual-layer Disentanglement Network for Shadow Detection
0 of 16 row(s) selected.