Visual instance retrieval with deep convolutional networks
2016 (English)In: ITE Transactions on Media Technology and Applications, ISSN 2186-7364, Vol. 4, no 3, 251-258 p.Article in journal (Refereed) Published
This paper provides an extensive study on the availability of image representations based on convolutional networks (ConvNets) for the task of visual instance retrieval. Besides the choice of convolutional layers, we present an efficient pipeline exploiting multi-scale schemes to extract local features, in particular, by taking geometric invariance into explicit account, i.e. positions, scales and spatial consistency. In our experiments using five standard image retrieval datasets, we demonstrate that generic ConvNet image representations can outperform other state-of-the-art methods if they are extracted appropriately.
Place, publisher, year, edition, pages
Institute of Image Information and Television Engineers , 2016. Vol. 4, no 3, 251-258 p.
Convolutional network, Learning representation, Multi-resolution search, Visual instance retrieval
Computer Vision and Robotics (Autonomous Systems)
IdentifiersURN: urn:nbn:se:kth:diva-195472ScopusID: 2-s2.0-84979503481OAI: oai:DiVA.org:kth-195472DiVA: diva2:1049836
QC 201611252016-11-252016-11-032016-12-09Bibliographically approved