Weakly Supervised Object Localization on PASCAL VOC 2007 (trainval)
88.1CorLoc (Aero)ML-LocNet-L+
Evaluation results
| Method | Links | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ML-LocNet-L+Model Type=multi-phase or ensemble2019.11 | 88.1 | — | 68.2 | 85.5 | 71.2 | 49.4 | 57.4 | 90.7 | 77.6 | 53.5 | 42.6 | 79.6 | 34.1 | 69.1 | 81.7 | 91.9 | 35.4 | 64.6 | 79.3 | 64.3 | 79.3 | 69.6 | |
| Multi-branch network (Ours)Orthogonal initialization=true, Surrounded Candidates Suppression=true, Refinement=Fast R-CNN2019.06 | 86.7 | 70 | — | 85.9 | 63.4 | 55.3 | 42 | 84.8 | 85.2 | 78.2 | 47.2 | 88.4 | 49 | 73.3 | 84 | 92.8 | 20.5 | 56.8 | 84.5 | 62.9 | 82.1 | 78.1 | |
| OICR-Ens.+FRCNNModel Type=multi-phase or ensemble, Refinement=Fast R-CNN2019.11 | 85.8 | — | 64.3 | 82.7 | 62.8 | 45.2 | 43.5 | 84.8 | 87 | 46.8 | 15.7 | 82.2 | 51 | 45.6 | 83.7 | 91.2 | 22.2 | 59.7 | 75.3 | 65.1 | 76.8 | 78.1 | |
| OICR2019.06 | 85.4 | 61.2 | — | 78 | 61.6 | 40.4 | 38.2 | 82.2 | 84.2 | 46.5 | 15.2 | 80.1 | 45.2 | 41.9 | 73.8 | 89.6 | 18.9 | 56 | 74.2 | 62.1 | 73 | 77.4 | |
| W2F+RPN+FSD2Model Type=multi-phase or ensemble2019.11 | 85.4 | — | 70.3 | 87.5 | 62.5 | 54.3 | 35.5 | 85.3 | 86.6 | 82.3 | 39.7 | 82.9 | 49.4 | 76.5 | 74.8 | 90 | 46.8 | 53.9 | 84.5 | 68.3 | 79.1 | 79.9 | |
| TS2C2019.06 | 84.2 | 61 | — | 74.1 | 61.3 | 52.1 | 32.1 | 76.7 | 82.9 | 66.6 | 42.3 | 70.6 | 39.5 | 57 | 61.2 | 89.6 | 9.3 | 54.6 | 72.2 | 60 | 65 | 70.3 | |
| Multi-branch network (Ours)Orthogonal initialization=true, Surrounded Candidates Suppression=true2019.06 | 84.2 | 66.9 | — | 84.7 | 59.5 | 52.7 | 37.8 | 81.2 | 83.3 | 72.4 | 41.6 | 84.9 | 43.7 | 69.5 | 75.9 | 90.8 | 18.1 | 54.9 | 81.4 | 60.8 | 79.1 | 80.6 | |
| TS2CModel Type=single end-to-end, Backbone=VGG162019.11 | 84.2 | — | 61 | 74.1 | 61.3 | 52.1 | 32.1 | 76.7 | 82.9 | 66.6 | 42.3 | 70.6 | 39.5 | 57 | 61.2 | 88.4 | 9.3 | 54.6 | 72.2 | 60 | 65 | 70.3 | |
| WCCN2019.06 | 83.9 | 56.7 | — | 72.8 | 64.5 | 44.1 | 40.1 | 65.7 | 82.5 | 58.9 | 33.7 | 72.5 | 25.6 | 53.7 | 67.4 | 88.4 | 26.8 | 49.1 | 68.1 | 27.9 | 64.5 | 55.7 | |
| WCCNModel Type=single end-to-end, Backbone=VGG162019.11 | 83.9 | — | 56.7 | 72.8 | 64.5 | 44.1 | 40.1 | 65.7 | 82.5 | 58.9 | 33.7 | 72.5 | 25.6 | 53.7 | 67.4 | 77.4 | 26.8 | 49.1 | 68.1 | 27.9 | 64.5 | 55.7 | |
| PCL-Ens.+FRCNNModel Type=multi-phase or ensemble, Refinement=Fast R-CNN2019.11 | 83.8 | — | 66.6 | 85.1 | 65.5 | 43.1 | 50.8 | 83.2 | 85.3 | 59.3 | 28.5 | 82.2 | 57.4 | 50.7 | 85 | 92 | 27.9 | 54.2 | 72.2 | 65.9 | 77.6 | 82.1 | |
| WSRPN-Ens.+FRCNNModel Type=multi-phase or ensemble, Refinement=Fast R-CNN2019.11 | 83.8 | — | 68.4 | 82.7 | 60.7 | 35.1 | 53.8 | 82.7 | 88.6 | 67.4 | 22 | 86.3 | 68.8 | 50.9 | 90.8 | 93.6 | 44 | 61.2 | 82.5 | 65.9 | 71.1 | 76.7 | |
| Kantorov et al.2017.04 | 83.3 | 55.1 | — | 68.6 | 54.7 | 23.4 | 18.3 | 73.6 | 74.1 | 54.1 | 8.6 | 65.1 | 47.1 | 59.5 | 67 | 83.5 | 35.3 | 39.9 | 67 | 49.7 | 63.5 | 65.2 | |
| ContextLocNetModel Type=single end-to-end, Backbone=VGG162019.11 | 83.3 | — | 55.1 | 68.6 | 54.7 | 23.4 | 18.3 | 73.6 | 74.1 | 54.1 | 8.6 | 65.1 | 47.1 | 59.5 | 67 | 83.5 | 35.3 | 39.9 | 67 | 49.7 | 63.5 | 65.2 | |
| Ours-Ens.Model Type=multi-phase or ensemble2019.11 | 83.3 | — | 70.6 | 85.5 | 68.8 | 56.9 | 49.6 | 84.3 | 87 | 83.1 | 44.2 | 86.3 | 55.5 | 54.4 | 81.6 | 92.8 | 22.8 | 60.4 | 81.4 | 70.2 | 81.4 | 81.4 | |
| PCL2019.06 | 81.7 | 63 | — | 82.4 | 63.4 | 41 | 42.4 | 79.7 | 84.2 | 54.9 | 23.4 | 78.8 | 54.4 | 46 | 75.9 | 89.6 | 22.8 | 51.3 | 72.2 | 66.1 | 74.9 | 76 | |
| OICRModel Type=single end-to-end, Backbone=VGG162019.11 | 81.7 | — | 60.6 | 80.4 | 48.7 | 49.5 | 32.8 | 81.7 | 85.4 | 40.1 | 40.6 | 79.5 | 35.7 | 33.7 | 60.5 | 88.8 | 21.8 | 57.9 | 76.3 | 59.9 | 75.3 | 81.4 | |
| MIL-OICR+GAM+REGModel Type=single end-to-end, Backbone=VGG162019.11 | 81.7 | — | 66.8 | 81.2 | 58.9 | 54.3 | 37.8 | 83.2 | 86.2 | 77 | 42.1 | 83.6 | 51.3 | 44.9 | 78.2 | 90.8 | 20.5 | 56.8 | 74.2 | 66.1 | 81 | 86 | |
| WSRPN2019.06 | 81.2 | 66.9 | — | 81.2 | 60.7 | 36.7 | 52.3 | 80.7 | 89 | 65.1 | 20.5 | 86.3 | 61.6 | 49.5 | 86.4 | 92.4 | 41.4 | 62.6 | 79.4 | 62.4 | 73 | 75.6 | |
| ZLDNModel Type=multi-phase or ensemble2019.11 | 80.3 | — | 61.5 | 76.5 | 64.2 | 40.9 | 46.7 | 78 | 84.3 | 57.6 | 21.1 | 69.5 | 28 | 46.8 | 70.7 | 89.4 | 41.9 | 54.7 | 76.3 | 61.1 | 76.3 | 65.2 | |
| Wang et al.2015.11 | 80.1 | 48.5 | — | 63.9 | 51.5 | 14.9 | 21 | 55.7 | 74.2 | 43.5 | 26.2 | 53.4 | 16.3 | 56.7 | 58.3 | 69.5 | 14.1 | 38.3 | 58.8 | 47.2 | 49.1 | 60.9 | |
| Wang et al.2017.04 | 80.1 | 48.5 | — | 63.9 | 51.5 | 14.9 | 21 | 55.7 | 74.2 | 43.5 | 26.2 | 53.4 | 16.3 | 56.7 | 58.3 | 69.5 | 14.1 | 38.3 | 58.8 | 47.2 | 49.1 | 60.9 | |
| Wang et al (2014)Neg.=true2017.07 | 80.1 | — | 47.7 | 63.9 | 51.5 | 14.9 | 21 | 55.7 | 74.2 | 43.5 | 26.2 | 53.4 | 16.3 | 56.7 | 58.3 | 69.5 | 14.1 | 38.3 | 58.8 | 47.2 | 49.1 | 60.9 | |
| MIL-PCL+GAM+REGModel Type=single end-to-end, Backbone=VGG162019.11 | 80 | — | 68 | 83.9 | 74.2 | 53.2 | 48.5 | 82.7 | 86.2 | 69.5 | 39.3 | 82.9 | 53.6 | 61.4 | 72.4 | 91.2 | 22.4 | 57.5 | 83.5 | 64.8 | 75.7 | 77.1 | |
| PCLModel Type=single end-to-end, Backbone=VGG162019.11 | 79.6 | — | 62.7 | 85.5 | 62.2 | 47.9 | 37 | 83.8 | 83.4 | 43 | 38.3 | 80.1 | 50.6 | 30.9 | 57.8 | 90.8 | 27 | 58.2 | 75.3 | 68.5 | 75.7 | 78.9 | |
| Ren et al (2016)Neg.=true2017.07 | 79.2 | — | 43.9 | 56.9 | 46 | 12.2 | 15.7 | 58.4 | 71.4 | 48.6 | 7.2 | 69.9 | 16.7 | 47.4 | 44.2 | 75.5 | 41.2 | 39.6 | 47.4 | 32.2 | 49.8 | 18.6 | |
| Li et al.2017.04 | 78.2 | 52.4 | — | 67.1 | 61.8 | 38.1 | 36.1 | 61.8 | 78.8 | 55.2 | 28.5 | 68.8 | 18.5 | 49.2 | 64.1 | 73.5 | 21.4 | 47.4 | 64.6 | 22.3 | 60.9 | 52.3 | |
| PDAModel Type=multi-phase or ensemble2019.11 | 78.2 | — | 52.4 | 67.1 | 61.8 | 38.1 | 36.1 | 61.8 | 78.8 | 55.2 | 28.5 | 68.8 | 18.5 | 49.2 | 64.1 | 73.5 | 21.4 | 47.4 | 64.6 | 22.3 | 60.9 | 52.3 | |
| WSRPNModel Type=single end-to-end, Backbone=VGG162019.11 | 77.5 | — | 63.8 | 81.2 | 55.3 | 19.7 | 44.3 | 80.2 | 86.6 | 69.5 | 10.1 | 87.7 | 68.4 | 52.1 | 84.4 | 91.6 | 57.4 | 63.4 | 77.3 | 58.1 | 57 | 53.8 | |
| HCP+DSD+OSSH2Dense Subgraph Discovery=true, Online Supportive Sample Harvesting=epochs 2-32017.04 | 73.9 | 54.9 | — | 56 | 52.1 | 26.9 | 34 | 66.6 | 80 | 59.5 | 13.1 | 70.2 | 22.9 | 55.7 | 60.6 | 83.8 | 22 | 51.5 | 71.1 | 50.4 | 71.2 | 74.4 | |
| HCP+DSD+OSSH3Dense Subgraph Discovery=true, Online Supportive Sample Harvesting=epochs 2-42017.04 | 72.7 | 56.1 | — | 55.3 | 53 | 27.8 | 35.2 | 68.6 | 81.9 | 60.7 | 11.6 | 71.6 | 29.7 | 54.3 | 64.3 | 88.2 | 22.2 | 53.7 | 72.2 | 52.6 | 68.9 | 75.5 | |
| Self-taughtModel Type=single end-to-end, Backbone=VGG162019.11 | 72.7 | — | 56.1 | 55.3 | 53 | 27.8 | 35.2 | 68.6 | 81.9 | 60.7 | 11.6 | 71.6 | 29.7 | 54.3 | 64.3 | 88.2 | 22.2 | 53.7 | 72.2 | 52.6 | 68.9 | 75.5 | |
| DDT+Neg.=false2017.07 | 71.4 | — | 48.5 | 65.6 | 64.6 | 25.5 | 8.5 | 64.8 | 61.3 | 80.5 | 10.3 | 49 | 26.5 | 72.6 | 75.2 | 69 | 9.9 | 12.2 | 39.7 | 55.7 | 75 | 32.5 | |
| HCP+DSD+OSSH1Dense Subgraph Discovery=true, Online Supportive Sample Harvesting=epoch 2 only2017.04 | 70.2 | 50.2 | — | 60 | 53.9 | 26.1 | 28.3 | 58.9 | 75.4 | 58.9 | 14.8 | 63.4 | 17.9 | 52.6 | 51.7 | 67 | 19.7 | 46.3 | 63.9 | 42.4 | 67 | 65.1 | |
| WSDDN Ensemblevariant=Ensemble2015.11 | 68.9 | 58 | — | 68.7 | 65.2 | 42.5 | 40.6 | 72.6 | 75.2 | 53.7 | 29.7 | 68.1 | 33.5 | 45.6 | 65.9 | 86.1 | 27.5 | 44.9 | 76 | 62.4 | 66.3 | 66.8 | |
| WSDDN2019.06 | 68.9 | 58 | — | 68.7 | 65.2 | 42.5 | 40.6 | 72.6 | 75.2 | 53.7 | 29.7 | 68.1 | 33.5 | 45.6 | 65.9 | 86.1 | 27.5 | 44.9 | 76 | 62.4 | 66.3 | 66.8 | |
| WSDDN-Ens.Model Type=multi-phase or ensemble2019.11 | 68.9 | — | 58 | 68.7 | 65.2 | 42.5 | 40.6 | 72.6 | 75.2 | 53.7 | 29.7 | 68.1 | 33.5 | 45.6 | 65.9 | 86.1 | 27.5 | 44.9 | 76 | 62.4 | 66.3 | 66.8 | |
| WSDDN Svariant=S2015.11 | 68.5 | 54.2 | — | 67.5 | 56.7 | 34.3 | 32.8 | 69.9 | 75 | 45.7 | 17.1 | 68.1 | 30.5 | 40.6 | 67.2 | 82.9 | 28.8 | 43.7 | 71.9 | 62 | 62.8 | 58.2 | |
| Shi et al (2013)Neg.=true2017.07 | 67.3 | — | 36.2 | 54.4 | 34.3 | 17.8 | 1.3 | 46.6 | 60.7 | 68.9 | 2.5 | 32.4 | 16.2 | 58.9 | 51.5 | 64.6 | 18.2 | 3.1 | 20.9 | 34.7 | 63.4 | 5.9 | |
| DDTNeg.=false2017.07 | 67.3 | — | 46.9 | 63.3 | 61.3 | 22.7 | 8.5 | 64.8 | 57 | 80.5 | 9.4 | 49 | 22.5 | 72.6 | 73.8 | 69 | 7.2 | 15 | 35.3 | 54.7 | 75 | 29.4 | |
| Bilen et al.2015.11 | 66.4 | 43.7 | — | 59.3 | 42.7 | 20.4 | 21.3 | 63.4 | 74.3 | 59.6 | 21.1 | 58.2 | 14 | 38.5 | 49.5 | 60 | 19.8 | 39.2 | 41.7 | 30.1 | 50.2 | 44.1 | |
| Bilen et al.2017.04 | 66.4 | 43.7 | — | 59.3 | 42.7 | 20.4 | 21.3 | 63.4 | 74.3 | 59.6 | 21.1 | 58.2 | 14 | 38.5 | 49.5 | 60 | 19.8 | 39.2 | 41.7 | 30.1 | 50.2 | 44.1 | |
| Bilen et al (2015)Neg.=true2017.07 | 66.4 | — | 43.7 | 59.3 | 42.7 | 20.4 | 21.3 | 63.4 | 74.3 | 59.6 | 21.1 | 58.2 | 14 | 38.5 | 49.5 | 60 | 19.8 | 39.2 | 41.7 | 30.1 | 50.2 | 44.1 | |
| Cinbis et al.2015.11 | 65.3 | 52 | — | 55 | 52.4 | 48.3 | 18.2 | 66.4 | 77.8 | 35.6 | 26.5 | 67 | 46.9 | 48.4 | 70.5 | 69.1 | 35.2 | 35.2 | 69.6 | 43.4 | 64.6 | 43.7 | |
| WSDDN Mvariant=M2015.11 | 65.1 | 56.1 | — | 63.4 | 59.7 | 45.9 | 38.5 | 69.4 | 77 | 50.7 | 30.1 | 68.8 | 34 | 37.3 | 61 | 82.9 | 25.1 | 42.9 | 79.2 | 59.4 | 68.2 | 64.1 | |
| WSDDN Lvariant=L2015.11 | 65.1 | 53.5 | — | 58.8 | 58.5 | 33.1 | 39.8 | 68.3 | 60.2 | 59.6 | 34.8 | 64.5 | 30.5 | 43 | 56.8 | 82.4 | 25.5 | 41.6 | 61.5 | 55.9 | 65.9 | 63.7 | |
| WSDDNModel Type=single end-to-end, Backbone=VGG162019.11 | 65.1 | — | 53.5 | 58.8 | 58.5 | 33.1 | 39.8 | 68.3 | 60.2 | 59.6 | 34.8 | 64.5 | 30.5 | 43 | 56.8 | 82.4 | 25.5 | 41.6 | 61.5 | 55.9 | 65.9 | 63.7 | |
| BaselineComponents=WSDDN + Refinement Module2019.06 | 64.2 | 64.8 | — | 83.5 | 63.1 | 45.2 | 38.5 | 82.2 | 86.7 | 57.6 | 35.5 | 83.6 | 41.8 | 69.5 | 69 | 90.4 | 20.1 | 56.8 | 83.5 | 66.9 | 78.3 | 79.6 | |
| Cinbis et al.2017.04 | 57.2 | 47.3 | — | 62.2 | 50.9 | 37.9 | 23.9 | 64.8 | 74.4 | 24.8 | 29.7 | 64.1 | 40.8 | 37.3 | 55.6 | 68.1 | 25.5 | 38.5 | 65.2 | 35.8 | 56.6 | 33.5 | |
| HCP+DSDDense Subgraph Discovery=true2017.04 | 56.9 | 37.9 | — | 36 | 45.4 | 26.5 | 13.2 | 49.8 | 54.5 | 53.1 | 15.9 | 45.6 | 13.4 | 37.5 | 38.1 | 42.1 | 16.2 | 34.2 | 45.4 | 29.7 | 55.6 | 46.1 | |
| Cinbis et al (2015)Neg.=true2017.07 | 56.6 | — | 38.8 | 58.3 | 28.4 | 20.7 | 6.8 | 54.9 | 69.1 | 20.8 | 9.2 | 50.5 | 10.2 | 29 | 58 | 64.9 | 36.7 | 18.7 | 56.5 | 13.2 | 54.9 | 59.4 | |
| HCPdescription=Baseline2017.04 | 54.4 | 34.1 | — | 37.2 | 42.1 | 28.1 | 15.7 | 13.8 | 47.8 | 49.6 | 40.6 | 16.4 | 38.7 | 13.8 | 34.5 | 22.2 | 10.8 | 36.4 | 42.3 | 20.8 | 46.1 | 49.3 | |
| Siva and Xiang (2011)Neg.=true2017.07 | 42.4 | — | 30.4 | 46.5 | 18.2 | 8.8 | 2.9 | 40.9 | 73.2 | 44.8 | 5.4 | 30.5 | 19 | 34 | 48.8 | 65.3 | 8.2 | 9.4 | 16.7 | 32.3 | 54.8 | 5.5 | |
| Wang et al (2015)Neg.=true2017.07 | 37.7 | — | 40.2 | 58.8 | 39 | 4.7 | 4 | 48.4 | 70 | 63.7 | 9 | 54.2 | 33.3 | 37.4 | 61.6 | 57.6 | 30.1 | 31.7 | 32.4 | 52.8 | 49 | 27.8 | |
| Bilen et al.2016.11 | — | — | 43.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cinbis et al.2016.11 | — | — | 52 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CL2019.06 | — | 64.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GAL-fWSDModel Type=multi-phase or ensemble2019.11 | — | — | 67.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Li et al.Backbone=AlexNet2016.11 | — | — | 49.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Li et al.Backbone=VGG162016.11 | — | — | 52.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MLEM2019.06 | — | 61.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OICRRefinement=Fast R-CNN2019.06 | — | 64.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCLRefinement=Fast R-CNN2019.06 | — | 66.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2016.11 | — | — | 48.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WCCN_2stageBackbone=AlexNet, Number of Stages=22016.11 | — | — | 51.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WCCN_2stageBackbone=VGG16, Number of Stages=22016.11 | — | — | 54.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WCCN_3stageBackbone=AlexNet, Number of Stages=32016.11 | — | — | 52.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WCCN_3stageBackbone=VGG16, Number of Stages=32016.11 | — | — | 56.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WSDDN2016.11 | — | — | 56.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WSRPNRefinement=Fast R-CNN2019.06 | — | 68.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ZLDN2019.06 | — | 61.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |