Salient Object Detection on HRSOD (test)
0.986F-betaSAM-UQ
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SAM-UQ2026.01 | 0.986 | 0.977 | 0.005 | — | — | — | 0.988 | — | — | — | |
| Pi-SAM2026.01 | 0.974 | 0.972 | 0.006 | — | — | — | 0.991 | — | — | — | |
| HQ-SAM2026.01 | 0.973 | 0.958 | 0.012 | — | — | — | 0.985 | — | — | — | |
| DIS-SAM2026.01 | 0.971 | 0.969 | 0.008 | — | — | — | 0.984 | — | — | — | |
| S3ODNetData=SOD2025.10 | 0.964 | 0.953 | 0.017 | — | — | — | 0.973 | — | — | — | |
| BiRefNetData=SOD2025.10 | 0.963 | 0.957 | 0.016 | — | — | — | 0.973 | — | — | — | |
| S3ODNetData=S3OD + SOD2025.10 | 0.963 | 0.961 | 0.013 | — | — | — | 0.978 | — | — | — | |
| BiRefNet2026.01 | 0.962 | 0.96 | 0.011 | — | — | — | 0.979 | — | — | — | |
| InSpyreNetData=SOD2025.10 | 0.956 | 0.956 | 0.018 | — | — | — | 0.962 | — | — | — | |
| SAM2026.01 | 0.955 | 0.932 | 0.022 | — | — | — | 0.963 | — | — | — | |
| S3ODNetTraining Data=S3OD2025.10 | 0.954 | 0.955 | 0.016 | — | — | — | 0.972 | — | — | — | |
| InSPyReNetbackbone=SwinB, training_datasets=DUTS-TR2022.09 | 0.949 | — | 0.016 | — | — | — | — | 0.952 | 0.738 | — | |
| PGNetTrained on=UHRSD-TR and HRSOD-TR, Backbone=ResNet-502022.04 | 0.945 | 0.938 | 0.02 | — | — | 57.147 | 0.946 | — | — | — | |
| PGNetData=SOD2025.10 | 0.945 | 0.938 | 0.02 | — | — | — | 0.946 | — | — | — | |
| PGNetbackbone=SwinB + ResNet18, training_datasets=HRSOD-TR, UHRSD-TR2022.09 | 0.939 | — | 0.02 | — | — | — | — | 0.938 | 0.727 | — | |
| PGNetTrained on=DUTS-TR and HRSOD-TR, Backbone=ResNet-502022.04 | 0.937 | 0.935 | 0.02 | — | — | 45.292 | 0.946 | — | — | — | |
| PGNetTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.931 | 0.93 | 0.021 | — | — | 46.923 | 0.944 | — | — | — | |
| PGNetbackbone=SwinB + ResNet18, training_datasets=DUTS-TR, HRSOD-TR2022.09 | 0.929 | — | 0.02 | — | — | — | — | 0.935 | 0.714 | — | |
| S3ODNetTraining Data=DIS2025.10 | 0.923 | 0.913 | 0.03 | — | — | — | 0.932 | — | — | — | |
| DHQSODTrained on=DUTS-TR and HRSOD-TR, Backbone=ResNet-502022.04 | 0.922 | 0.92 | 0.022 | — | — | 46.495 | 0.947 | — | — | — | |
| PGNetbackbone=SwinB + ResNet18, training_datasets=DUTS-TR2022.09 | 0.922 | — | 0.021 | — | — | — | — | 0.93 | 0.693 | — | |
| DHQ2026.01 | 0.922 | 0.92 | 0.022 | — | — | — | 0.947 | — | — | — | |
| Ours-DHTraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.922 | 0.922 | 0.022 | — | — | 14.266 | — | — | — | 0.571 | |
| DHQData=SOD2025.10 | 0.922 | 0.92 | 0.022 | — | — | — | 0.947 | — | — | — | |
| Ours-DHTraining datasets=DUTS-TR + HRSOD-Training (resized), Backbone=VGG-162021.08 | 0.921 | 0.917 | 0.024 | — | — | 14.412 | — | — | — | 0.531 | |
| PA-KRNbackbone=SwinB, training_datasets=DUTS-TR, re-implemented=true2022.09 | 0.918 | — | 0.026 | — | — | — | — | 0.927 | 0.653 | — | |
| OursTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.918 | 0.912 | 0.027 | — | — | 15.676 | — | — | — | 0.536 | |
| MINetbackbone=SwinB, training_datasets=DUTS-TR, re-implemented=true2022.09 | 0.917 | — | 0.025 | — | — | — | — | 0.927 | 0.67 | — | |
| Tang et al.backbone=ResNet50, training_datasets=DUTS-TR, HRSOD-TR2022.09 | 0.915 | — | 0.022 | — | — | — | — | 0.92 | 0.693 | — | |
| OursTraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.915 | 0.919 | 0.024 | — | — | 14.396 | — | — | — | 0.576 | |
| PFSTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.911 | 0.906 | 0.033 | — | — | 63.537 | 0.922 | — | — | — | |
| Chen et al.backbone=SwinB, training_datasets=DUTS-TR, re-implemented=true2022.09 | 0.907 | — | 0.032 | — | — | — | — | 0.915 | 0.684 | — | |
| CTDTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.905 | 0.905 | 0.032 | — | — | 63.907 | 0.921 | — | — | — | |
| HRSODTrained on=DUTS-TR and HRSOD-TR, Backbone=VGG162022.04 | 0.905 | 0.896 | 0.03 | — | — | 88.017 | 0.934 | — | — | — | |
| HRSOD2026.01 | 0.905 | 0.896 | 0.03 | — | — | — | 0.934 | — | — | — | |
| GateNetTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.905 | 0.906 | 0.035 | — | — | 79.468 | — | — | — | 0.886 | |
| HRNetTraining datasets=DUTS-TR + HRSOD-Training, Backbone=HRNet2021.08 | 0.905 | 0.897 | 0.03 | — | — | 88.017 | — | — | — | 0.888 | |
| CSFTraining datasets=DUTS-TR, Backbone=Res2Net502021.08 | 0.905 | 0.905 | 0.032 | — | — | 58.655 | — | — | — | 0.812 | |
| HRSODData=SOD2025.10 | 0.905 | 0.896 | 0.03 | — | — | — | 0.934 | — | — | — | |
| LDFTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.904 | 0.904 | 0.032 | — | — | 58.714 | 0.919 | — | — | — | |
| LDF2026.01 | 0.904 | 0.904 | 0.032 | — | — | — | 0.919 | — | — | — | |
| LDFData=SOD2025.10 | 0.904 | 0.904 | 0.032 | — | — | — | 0.919 | — | — | — | |
| F³Netbackbone=SwinB, training_datasets=DUTS-TR, re-implemented=true2022.09 | 0.902 | — | 0.034 | — | — | — | — | 0.912 | 0.674 | — | |
| MINetTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.902 | 0.903 | 0.032 | — | — | 76.291 | — | — | — | 0.849 | |
| MVANetTraining Data=DIS2025.10 | 0.902 | 0.919 | 0.033 | — | — | — | 0.93 | — | — | — | |
| F3NetTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.9 | 0.897 | 0.035 | — | — | 65.757 | 0.913 | — | — | — | |
| LDFTraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.9 | 0.897 | 0.035 | — | — | 65.901 | — | — | — | 0.817 | |
| ITSDTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.896 | 0.898 | 0.036 | — | — | 87.946 | 0.912 | — | — | — | |
| PFPN+Training datasets=DUTS-TR, Backbone=ResNet-50, Post-processed by CRF=true2021.08 | 0.894 | 0.9 | 0.038 | — | — | 71.293 | — | — | — | 0.922 | |
| DASNetTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.893 | 0.897 | 0.032 | — | — | 69.31 | 0.925 | — | — | — | |
| Zeng et al.backbone=VGG16, training_datasets=DUTS-TR, HRSOD-TR2022.09 | 0.892 | — | 0.03 | — | — | — | — | 0.897 | 0.623 | — | |
| InSpyreNetTraining Data=DIS2025.10 | 0.891 | 0.912 | 0.038 | — | — | — | 0.923 | — | — | — | |
| GCPATrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.889 | 0.898 | 0.036 | — | — | 74.9 | 0.898 | — | — | — | |
| GCPATraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.889 | 0.897 | 0.042 | — | — | 65.048 | — | — | — | 0.896 | |
| F3NTraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.889 | 0.894 | 0.039 | — | — | 70.32 | — | — | — | 0.873 | |
| GLFNVariant=Ours-DH2019.08 | 0.888 | 0.897 | 0.03 | — | — | — | — | — | — | — | |
| LDFbackbone=SwinB, training_datasets=DUTS-TR, re-implemented=true2022.09 | 0.888 | — | 0.036 | — | — | — | — | 0.905 | 0.672 | — | |
| BiRefNetTraining Data=DIS2025.10 | 0.887 | 0.915 | 0.031 | — | — | — | 0.926 | — | — | — | |
| EGNetTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.883 | 0.888 | 0.044 | — | — | 73.5 | — | — | — | 0.896 | |
| SCRNTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.88 | 0.888 | 0.042 | — | — | 75.696 | 0.887 | — | — | — | |
| BASNetTraining datasets=DUTS-TR, Backbone=ResNet-502021.08 | 0.878 | 0.89 | 0.038 | — | — | 67.643 | — | — | — | 0.823 | |
| CPDTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.876 | 0.887 | 0.039 | — | — | 72.686 | — | — | — | 0.824 | |
| CPDTrained on=DUTS-TR, Backbone=ResNet-502022.04 | 0.867 | 0.881 | 0.041 | — | — | 62.066 | 0.891 | — | — | — | |
| GLFNVariant=Ours-D2019.08 | 0.857 | 0.876 | 0.04 | — | — | — | — | — | — | — | |
| DSS+Training datasets=MSRA-B, Backbone=VGG-162021.08 | 0.826 | 0.84 | 0.06 | — | — | 145.403 | — | — | — | 0.952 | |
| ITSDTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.824 | 0.834 | 0.071 | — | — | 139.943 | — | — | — | 0.924 | |
| DGRLTraining datasets=DUTS-TR, Backbone=VGG-162021.08 | 0.821 | 0.847 | 0.055 | — | — | 95.034 | — | — | — | 0.889 | |
| AmuletTraining datasets=MSRA10K, Backbone=VGG-162021.08 | 0.799 | 0.829 | 0.075 | — | — | 139.889 | — | — | — | 0.947 | |
| R3NetTraining datasets=MSRA10K, Backbone=ResNeXt-1012021.08 | 0.798 | 0.812 | 0.081 | — | — | 108.91 | — | — | — | 0.931 | |
| DGF2019.08 | 0.795 | 0.824 | 0.058 | — | — | — | — | — | — | — | |
| DGRL2019.08 | 0.789 | 0.848 | 0.053 | — | — | — | — | — | — | — | |
| RAS2019.08 | 0.773 | 0.842 | 0.058 | — | — | — | — | — | — | — | |
| NLDF2019.08 | 0.763 | 0.853 | 0.055 | — | — | — | — | — | — | — | |
| DSS2019.08 | 0.756 | 0.84 | 0.06 | — | — | — | — | — | — | — | |
| DHS2019.08 | 0.746 | 0.848 | 0.059 | — | — | — | — | — | — | — | |
| Amulet2019.08 | 0.717 | 0.83 | 0.075 | — | — | — | — | — | — | — | |
| UCF2019.08 | 0.7 | 0.819 | 0.095 | — | — | — | — | — | — | — | |
| RFCN2019.08 | 0.53 | 0.608 | 0.121 | — | — | — | — | — | — | — | |
| AmuletInput Resolution=384 x 3842019.08 | — | — | — | 0.05 | 132.6 | — | — | — | — | — | |
| Amulet2019.08 | — | — | — | — | — | 25.75 | — | — | — | — | |
| DGFInput Resolution=384 x 3842019.08 | — | — | — | 0.41 | 248.9 | — | — | — | — | — | |
| DGF2019.08 | — | — | — | — | — | 32.91 | — | — | — | — | |
| DGRLInput Resolution=384 x 3842019.08 | — | — | — | 0.52 | 648 | — | — | — | — | — | |
| DGRL2019.08 | — | — | — | — | — | 30.1 | — | — | — | — | |
| DHSInput Resolution=384 x 3842019.08 | — | — | — | 0.05 | 376.2 | — | — | — | — | — | |
| DHS2019.08 | — | — | — | — | — | 22.85 | — | — | — | — | |
| DSSInput Resolution=384 x 3842019.08 | — | — | — | 5.12 | 447.3 | — | — | — | — | — | |
| DSS2019.08 | — | — | — | — | — | 25.53 | — | — | — | — | |
| GLFNInput Resolution=1024 x 10242019.08 | — | — | — | 0.39 | 129.6 | — | — | — | — | — | |
| GLFNInput Resolution=384 x 3842019.08 | — | — | — | 0.05 | 129.6 | — | — | — | — | — | |
| NLDFInput Resolution=384 x 3842019.08 | — | — | — | 2.31 | 425.9 | — | — | — | — | — | |
| NLDF2019.08 | — | — | — | — | — | 22.34 | — | — | — | — | |
| Ours2019.08 | — | — | — | — | — | 17.57 | — | — | — | — | |
| RASInput Resolution=384 x 3842019.08 | — | — | — | 0.08 | 81 | — | — | — | — | — | |
| RAS2019.08 | — | — | — | — | — | 26.26 | — | — | — | — | |
| RFCNInput Resolution=384 x 3842019.08 | — | — | — | 4.54 | 1,126.4 | — | — | — | — | — | |
| RFCN2019.08 | — | — | — | — | — | 68.98 | — | — | — | — | |
| UCFInput Resolution=384 x 3842019.08 | — | — | — | 0.14 | 117.9 | — | — | — | — | — | |
| UCF2019.08 | — | — | — | — | — | 22.84 | — | — | — | — |