Skip to Main content Skip to Navigation
Journal articles

NetVLAD: CNN architecture for weakly supervised place recognition

Abstract : We tackle the problem of large scale visual place recognition, where the task is to quickly and accurately recognize the location of a given query photograph. We present the following four principal contributions. First, we develop a convolutional neural network (CNN) architecture that is trainable in an end-to-end manner directly for the place recognition task. The main component of this architecture, NetVLAD, is a new generalized VLAD layer, inspired by the " Vector of Locally Aggregated Descriptors " image representation commonly used in image retrieval. The layer is readily pluggable into any CNN architecture and amenable to training via backpropagation. Second, we create a new weakly supervised ranking loss, which enables end-to-end learning of the architecture's parameters from images depicting the same places over time downloaded from Google Street View Time Machine. Third, we develop an efficient training procedure which can be applied on very large-scale weakly labelled tasks. Finally, we show that the proposed architecture and training procedure significantly outperform non-learnt image representations and off-the-shelf CNN descriptors on challenging place recognition and image retrieval benchmarks.
Document type :
Journal articles
Complete list of metadata

Cited literature [105 references]  Display  Hide  Download
Contributor : Relja Arandjelović Connect in order to contact the contributor
Submitted on : Thursday, July 6, 2017 - 1:06:39 AM
Last modification on : Tuesday, July 5, 2022 - 8:38:31 AM
Long-term archiving on: : Tuesday, January 23, 2018 - 9:48:36 PM


Files produced by the author(s)


  • HAL Id : hal-01557234, version 1



Relja Arandjelovic, Petr Gronat, Akihiko Torii, Tomas Pajdla, Josef Sivic. NetVLAD: CNN architecture for weakly supervised place recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, Institute of Electrical and Electronics Engineers, 2017, XX. ⟨hal-01557234⟩



Record views


Files downloads