Image Retrieval on Holidays standard (test)
94.8mAPR-MAC
Evaluation Results
| Method | Links | |
|---|---|---|
| R-MACNet=Res, F-tuned=yes, Dim=20482017.11 | 94.8 | |
| GeMNet=Res, F-tuned=yes, Dim=20482017.11 | 93.9 | |
| R-MACNet=Res, F-tuned=no, Dim=20482017.11 | 91.3 | |
| PlaNetaugmentation=true, spatial search=true2016.02 | 89.9 | |
| GeMNet=VGG, F-tuned=yes, Dim=5122017.11 | 89.5 | |
| R-MACNet=VGG, F-tuned=yes, Dim=5122017.11 | 89.1 | |
| NetVLADNet=VGG, F-tuned=no, Dim=40962017.11 | 88.3 | |
| NetVLADNet=VGG, F-tuned=yes, Dim=40962017.11 | 87.5 | |
| R-MACNet=VGG, F-tuned=no, Dim=5122017.11 | 86.9 | |
| NetVLADNet=VGG, F-tuned=yes, Dim=5122017.11 | 86.1 | |
| NetVladDimensionality=256, Backbone=VGG, Training Protocol=Off-the-shelf2016.04 | 86 | |
| CroWNet=VGG, F-tuned=no, Dim=5122017.11 | 85.1 | |
| Overfeataugmentation=true, spatial search=true2016.02 | 84.3 | |
| SPOCNet=VGG, F-tuned=no, Dim=5122017.11 | 83.9 | |
| ASMK+MA2016.02 | 82.2 | |
| MACDimensionality=512, Backbone=VGG, Training Protocol=Fine-tuned2016.04 | 79.5 | |
| AlexNet+LM Retraining2016.02 | 79.3 | |
| MACNet=VGG, F-tuned=no, Dim=5122017.11 | 79 | |
| Hamming Embedding2016.02 | 77.5 | |
| BoW-fVocab+R+QENet=n/a, F-tuned=n/a, Dim=n/a2017.11 | 75.8 | |
| Fine Vocabulary2016.02 | 74.9 | |
| PlaNet2016.02 | 73.3 | |
| Overfeat2016.02 | 64.2 | |
| BoVW2016.02 | 57.2 | |
| GIST2016.02 | 37.6 |