Keras implementation of R-MAC+ descriptors.
The image below represents the query phase exeucted for the R-MAC+ descriptors.
- Keras (> 2.0.0)
- Tensorflow (> 1.5)
- Scipy
- Sklearn
- OpenCV 3
The pipeline was tested with VGG16 and ResNet50. For the VGG16 the best performance are reached when the features are extracted from the block5_pool, instead for ResNet from the activation_43. It is possible to try with other networks. Please before to try it, check if there are available the Keras weight for the selected network.
- Holidays
- Oxford5k
- Paris6k
Download the datasets and put it into the data folder. Then compile the script for the evaluation of the retrieval system.
python3 Keras_test_MAC.py
| Method | Network | Oxford5k | Paris6k | Holidays |
|---|---|---|---|---|
| R-MAC | VGG16 | 65.56% | 82.80% | 87.65% |
| R-MAC | ResNet50 | 71.77% | 83.31% | 92.55% |
| M-R RMAC+ | ResNet50 | 78.88% | 88.63% | 94.63% / 95.58% |
| M-R RMAC+ with retrieval based on 'db regions' | ResNet50 | 85.39 % | 91.90% | 94.37% / 95.87% |
The R-MAC is an our re-implementation of the Tolias et al. 2016 paper, instead M-R RMAC comes from the Gordo et al. 2016 paper. The last two experiments are also executed on the rotated version of Holidays.
@article{magliani2018accurate,
title={An accurate retrieval through R-MAC+ descriptors for landmark recognition},
author={Magliani, Federico and Prati, Andrea},
journal={arXiv preprint arXiv:1806.08565},
year={2018}
}
@article{tolias2015particular,
title={Particular object retrieval with integral max-pooling of CNN activations},
author={Tolias, Giorgos and Sicre, Ronan and J{\'e}gou, Herv{\'e}},
journal={arXiv preprint arXiv:1511.05879},
year={2015}
}
@inproceedings{gordo2016deep,
title={Deep image retrieval: Learning global representations for image search},
author={Gordo, Albert and Almaz{\'a}n, Jon and Revaud, Jerome and Larlus, Diane},
booktitle={European Conference on Computer Vision},
pages={241--257},
year={2016},
organization={Springer}
}