Video saliency prediction on DHF1K (test)
0.908AUC-JHD2SDSL
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| HD2SDSLadaptation=Domain-Specific Learning2020.10 | 0.908 | 0.406 | 0.702 | 0.503 | 2.812 | |
| ViNetFPS=62.52020.12 | 0.908 | 0.381 | 0.728 | 0.51 | 2.87 | |
| HD2SDAadaptation=Hierarchical Domain Adaptation2020.10 | 0.902 | 0.381 | 0.709 | 0.491 | 2.709 | |
| UNISALTraining setting=(v)2020.03 | 0.901 | 0.384 | 0.692 | 0.488 | 2.739 | |
| UNISALTraining setting=(vi)2020.03 | 0.901 | 0.39 | 0.691 | 0.49 | 2.776 | |
| UNISAL2020.10 | 0.901 | 0.39 | 0.691 | 0.49 | 2.776 | |
| HD2Sadaptation=None2020.10 | 0.901 | 0.406 | 0.699 | 0.497 | 2.781 | |
| UNISALFPS=1112020.12 | 0.901 | 0.39 | 0.691 | 0.49 | 2.77 | |
| UNISALTraining setting=(i)2020.03 | 0.899 | 0.378 | 0.686 | 0.481 | 2.707 | |
| SalSAC2020.10 | 0.896 | 0.357 | 0.697 | 0.479 | 2.673 | |
| SALSACFPS=502020.12 | 0.896 | 0.357 | 0.697 | 0.479 | 2.67 | |
| TASED-Net2019.08 | 0.895 | 0.361 | 0.712 | 0.47 | 2.667 | |
| TASED-NetModel Category=Dynamic model2020.03 | 0.895 | 0.361 | 0.712 | 0.47 | 2.667 | |
| STRA-NetModel Category=Dynamic model2020.03 | 0.895 | 0.355 | 0.663 | 0.458 | 2.558 | |
| SalEMAModel Category=Dynamic model, Training setting=(vi)2020.03 | 0.895 | 0.283 | 0.739 | 0.414 | 2.285 | |
| STRA-Net2020.10 | 0.895 | 0.355 | 0.663 | 0.458 | 2.558 | |
| TASED-Net2020.10 | 0.895 | 0.361 | 0.712 | 0.47 | 2.667 | |
| STRA-NetFPS=502020.12 | 0.895 | 0.355 | 0.663 | 0.458 | 2.55 | |
| TASED-NetFPS=16.62020.12 | 0.895 | 0.361 | 0.712 | 0.47 | 2.66 | |
| Attentive CNN-LSTMTraining Setting=(iv) DHF1K+Hollywood-2+UCF sports2018.01 | 0.89 | 0.315 | 0.601 | 0.434 | 2.354 | |
| SalEMA2019.07 | 0.89 | 0.465 | 0.667 | 0.449 | 2.573 | |
| ACLnet2019.07 | 0.89 | 0.315 | 0.601 | 0.434 | 2.354 | |
| ACLNet2019.08 | 0.89 | 0.315 | 0.601 | 0.434 | 2.354 | |
| ACLNetModel Category=Dynamic model, Training setting=(vi)2020.03 | 0.89 | 0.315 | 0.601 | 0.434 | 2.354 | |
| SalEMAModel Category=Dynamic model, Fine-tuned=true2020.03 | 0.89 | 0.465 | 0.667 | 0.449 | 2.573 | |
| ACLNet2020.10 | 0.89 | 0.315 | 0.601 | 0.434 | 2.354 | |
| SalEMA2020.10 | 0.89 | 0.466 | 0.667 | 0.449 | 2.574 | |
| SALEMAFPS=1002020.12 | 0.89 | 0.466 | 0.667 | 0.449 | 2.57 | |
| ACLNetFPS=502020.12 | 0.89 | 0.315 | 0.601 | 0.434 | 2.35 | |
| SalCLSTM2019.07 | 0.887 | 0.322 | 0.693 | 0.435 | 2.364 | |
| Attentive CNN-LSTMTraining Setting=(i) DHF1K2018.01 | 0.885 | 0.311 | 0.553 | 0.415 | 2.259 | |
| DVAModel Category=Static model, Training setting=(vi)2020.03 | 0.883 | 0.297 | 0.623 | 0.397 | 2.237 | |
| UNISALTraining setting=(iv)2020.03 | 0.883 | 0.288 | 0.715 | 0.41 | 2.259 | |
| UNISALTraining setting=(ii)2020.03 | 0.881 | 0.313 | 0.69 | 0.422 | 2.352 | |
| Attentive CNN-LSTMTraining Setting=(ii) Hollywood-22018.01 | 0.878 | 0.297 | 0.543 | 0.388 | 2.125 | |
| Deep-NetModel Category=Static model, Training setting=(vi)2020.03 | 0.874 | 0.288 | 0.61 | 0.374 | 1.983 | |
| Attentive CNN-LSTM (Attention module ablation)Training Setting=(iv) DHF1K+Hollywood-2+UCF sports2018.01 | 0.87 | 0.273 | 0.577 | 0.38 | 2.077 | |
| UNISALTraining setting=(iii)2020.03 | 0.869 | 0.286 | 0.664 | 0.375 | 2.056 | |
| Attentive CNN-LSTMTraining Setting=(iii) UCF sports2018.01 | 0.866 | 0.277 | 0.596 | 0.362 | 1.951 | |
| SalGAN2019.07 | 0.866 | 0.262 | 0.709 | 0.37 | 2.043 | |
| SalGAN2019.08 | 0.866 | 0.262 | 0.709 | 0.37 | 2.043 | |
| SalGANModel Category=Static model2020.03 | 0.866 | 0.262 | 0.709 | 0.37 | 2.043 | |
| SalGAN2020.10 | 0.866 | 0.262 | 0.709 | 0.37 | 2.043 | |
| DVAModel Type=Static, Deep Learning=true2018.01 | 0.86 | 0.262 | 0.595 | 0.358 | 2.013 | |
| DVA2019.07 | 0.86 | 0.262 | 0.595 | 0.358 | 2.013 | |
| DVA2019.08 | 0.86 | 0.262 | 0.595 | 0.358 | 2.013 | |
| DVAModel Category=Static model2020.03 | 0.86 | 0.262 | 0.595 | 0.358 | 2.013 | |
| DVA2020.10 | 0.86 | 0.262 | 0.595 | 0.358 | 2.013 | |
| SALICONModel Type=Static, Deep Learning=true2018.01 | 0.857 | 0.232 | 0.59 | 0.327 | 1.901 | |
| SALICON2019.08 | 0.857 | 0.232 | 0.59 | 0.327 | 1.901 | |
| SALICONModel Category=Static model2020.03 | 0.857 | 0.232 | 0.59 | 0.327 | 1.901 | |
| SALICON2020.10 | 0.857 | 0.232 | 0.59 | 0.327 | 1.901 | |
| OM-CNNModel Type=Dynamic, Deep Learning=true2018.01 | 0.856 | 0.256 | 0.583 | 0.344 | 1.911 | |
| OM-CNN2019.08 | 0.856 | 0.256 | 0.583 | 0.344 | 1.911 | |
| SalNetModel Category=Dynamic model2020.03 | 0.856 | 0.256 | 0.583 | 0.344 | 1.911 | |
| OM-CNN2020.10 | 0.856 | 0.256 | 0.583 | 0.344 | 1.911 | |
| Deep-NetModel Type=Static, Deep Learning=true2018.01 | 0.855 | 0.201 | 0.592 | 0.331 | 1.775 | |
| Attentive CNN-LSTM (Attention module ablation)Training Setting=(ii) Hollywood-22018.01 | 0.855 | 0.25 | 0.541 | 0.318 | 1.703 | |
| Deep Net2019.08 | 0.855 | 0.201 | 0.592 | 0.331 | 1.775 | |
| Deep-NetModel Category=Static model2020.03 | 0.855 | 0.201 | 0.592 | 0.331 | 1.775 | |
| Deep Net2020.10 | 0.855 | 0.201 | 0.592 | 0.331 | 1.775 | |
| Attentive CNN-LSTM (Attention module ablation)Training Setting=(i) DHF1K2018.01 | 0.854 | 0.251 | 0.545 | 0.332 | 1.755 | |
| Attentive CNN-LSTM (Attention module ablation)Training Setting=(iii) UCF sports2018.01 | 0.852 | 0.26 | 0.582 | 0.35 | 1.945 | |
| Two-streamModel Type=Dynamic, Deep Learning=true2018.01 | 0.834 | 0.197 | 0.581 | 0.325 | 1.632 | |
| STSConvNet2019.08 | 0.834 | 0.197 | 0.581 | 0.325 | 1.632 | |
| OM-CNNModel Category=Dynamic model2020.03 | 0.834 | 0.197 | 0.581 | 0.325 | 1.632 | |
| STSConvNet2020.10 | 0.834 | 0.197 | 0.581 | 0.325 | 1.632 | |
| Shallow-NetModel Type=Static, Deep Learning=true2018.01 | 0.833 | 0.182 | 0.529 | 0.295 | 1.509 | |
| Shallow-NetModel Category=Static model2020.03 | 0.833 | 0.182 | 0.529 | 0.295 | 1.509 | |
| GBVSModel Type=Static, Deep Learning=false2018.01 | 0.828 | 0.186 | 0.554 | 0.283 | 1.474 | |
| GBVS2019.08 | 0.828 | 0.186 | 0.554 | 0.283 | 1.474 | |
| GBVSModel Category=Static model2020.03 | 0.828 | 0.186 | 0.554 | 0.283 | 1.474 | |
| GBVS2020.10 | 0.828 | 0.186 | 0.554 | 0.283 | 1.474 | |
| Fang et al.Model Type=Dynamic, Deep Learning=false2018.01 | 0.819 | 0.198 | 0.537 | 0.273 | 1.539 | |
| Fang et al.Model Category=Dynamic model2020.03 | 0.819 | 0.198 | 0.537 | 0.273 | 1.539 | |
| ITTIModel Type=Static, Deep Learning=false2018.01 | 0.774 | 0.162 | 0.553 | 0.233 | 1.207 | |
| ITTIModel Category=Static model2020.03 | 0.774 | 0.162 | 0.553 | 0.233 | 1.207 | |
| Rudoy et al.Model Type=Dynamic, Deep Learning=false2018.01 | 0.769 | 0.214 | 0.501 | 0.285 | 1.498 | |
| Rudoy et al.Model Category=Dynamic model2020.03 | 0.769 | 0.214 | 0.501 | 0.285 | 1.498 | |
| Hou et al.Model Type=Dynamic, Deep Learning=false2018.01 | 0.726 | 0.167 | 0.545 | 0.15 | 0.847 | |
| Hou et al.Model Category=Dynamic model2020.03 | 0.726 | 0.167 | 0.545 | 0.15 | 0.847 | |
| AWS-DModel Type=Dynamic, Deep Learning=false2018.01 | 0.703 | 0.157 | 0.513 | 0.174 | 0.94 | |
| AWS-DModel Category=Dynamic model2020.03 | 0.703 | 0.157 | 0.513 | 0.174 | 0.94 | |
| PQFTModel Type=Dynamic, Deep Learning=false2018.01 | 0.699 | 0.139 | 0.562 | 0.137 | 0.749 | |
| PQFTModel Category=Dynamic model2020.03 | 0.699 | 0.139 | 0.562 | 0.137 | 0.749 | |
| OBDLModel Type=Dynamic, Deep Learning=false2018.01 | 0.638 | 0.171 | 0.5 | 0.117 | 0.495 | |
| OBDLModel Category=Dynamic model2020.03 | 0.638 | 0.171 | 0.5 | 0.117 | 0.495 | |
| Seo et al.Model Type=Dynamic, Deep Learning=false2018.01 | 0.635 | 0.142 | 0.499 | 0.07 | 0.334 | |
| Seo et al.Model Category=Dynamic model2020.03 | 0.635 | 0.142 | 0.499 | 0.07 | 0.334 |