Dichotomous Image Segmentation On Dis Te4

評価指標

E-measure

HCE

MAE

S-Measure

max F-Measure

weighted F-measure

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	E-measure	HCE	MAE	S-Measure	max F-Measure	weighted F-measure	Paper Title	Repository
HySM	0.842	3331	0.091	0.802	0.782	0.693	HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation
U2Net	0.847	3653	0.087	0.807	0.795	0.705	U$^2$-Net: Going Deeper with Nested U-Structure for Salient Object Detection
MBV3	0.848	3817	0.098	0.770	0.736	0.664	Searching for MobileNetV3
HRNet	0.854	3864	0.092	0.792	0.772	0.687	Deep High-Resolution Representation Learning for Visual Recognition
InSPyReNet	0.926	2243	0.046	0.891	0.892	0.840	Revisiting Image Pyramid Structure for High Resolution Salient Object Detection
ICNet	0.837	3690	0.099	0.776	0.749	0.663	ICNet for Real-Time Semantic Segmentation on High-Resolution Images
BASNet	0.848	3601	0.091	0.794	0.780	0.693	BASNet: Boundary-Aware Salient Object Detection	-
IS-Net	0.87	2888	0.072	0.83	0.827	0.753	Highly Accurate Dichotomous Image Segmentation
InSPyReNet (HR scale)	0.926	2336	0.042	0.905	0.905	0.848	Revisiting Image Pyramid Structure for High Resolution Salient Object Detection
UNet	0.821	3218	0.102	-	0.759	0.659	U-Net: Convolutional Networks for Biomedical Image Segmentation
PFNet	0.838	3803	0.107	0.763	0.731	0.647	Camouflaged Object Segmentation with Distraction Mining
GCPANet	0.767	3678	0.127	0.723	0.670	0.559	Global Context-Aware Progressive Aggregation Network for Salient Object Detection
SINetV2	0.824	3683	0.113	0.744	0.699	0.616	Concealed Object Detection
DeeplabV3+	0.820	3709	0.111	0.744	0.715	0.621	Rethinking Atrous Convolution for Semantic Image Segmentation
PSPNet	0.815	3806	0.107	0.758	0.725	0.630	Pyramid Scene Parsing Network
BSV1	0.788	3999	0.114	0.755	0.710	0.598	BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
STDC	0.841	3819	0.102	0.762	0.731	0.652	Rethinking BiSeNet For Real-time Semantic Segmentation
F3Net	0.825	3760	0.107	0.752	0.721	0.633	F3Net: Fusion, Feedback and Focus for Salient Object Detection
BiRefNet	0.937	2746	0.040	0.898	0.900	0.861	Bilateral Reference for High-Resolution Dichotomous Image Segmentation
GateNet	0.803	3654	0.109	0.743	0.729	0.625	Suppress and Balance: A Simple Gated Network for Salient Object Detection

0 of 22 row(s) selected.