Saliency weighted convolutional features for instance search
Mohedano, Eva, McGuinness, KevinORCID: 0000-0003-1336-6477, Giró-i-Nieto, Xavier and O'Connor, Noel E.ORCID: 0000-0002-4033-9135
(2018)
Saliency weighted convolutional features for instance search.
In: 16th International Conference on Content-Based Multimedia Indexing 2018, 4 - 6 Sept 2018, La Rochelle, France.
ISBN 978-1-5386-7021-7
This work explores attention models to weight the contribution of local convolutional representations for the instance search task. We present a retrieval framework based on bags of local convolutional features (BLCF) that benefits from saliency weighting to build an efficient image representation. The use of human visual attention models (saliency) allows significant improvements in retrieval performance without the need to conduct region analysis or spatial verification, and without requiring any feature fine tuning. We investigate the impact of different saliency models, finding that higher performance on saliency benchmarks does not necessarily equate to improved performance when used in instance search tasks. The proposed approach outperforms the state-of-the-art on the challenging INSTRE benchmark by a large margin, and provides similar performance on the Oxford and Paris benchmarks compared to more complex methods that use off-the-shelf representations.Source code is publicly available at https://github.com/imatge-upc/salbow.
Metadata
Item Type:
Conference or Workshop Item (Paper)
Event Type:
Conference
Refereed:
Yes
Uncontrolled Keywords:
Instance Retrieval; Convolutional Neural Networks;
Bag of Words; Saliency weighting
This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
Funders:
Science Foundation Ireland SFI/12/RC/2289, Science Foundation Ireland SFI/15/SIRG/3283, Spanish Ministry of Economy and Competitivity & European Regional Development Fund (ERDF) under TEC2016-75976-R.
ID Code:
22456
Deposited On:
23 Jul 2018 11:40 by
Eva Mohedano Robles
. Last Modified 25 Jan 2019 10:03