Leveraging Implicit Spatial Information In Global Features For Image Retrieval

Pierre Jacob (1) , David Picard (2, 3, 4) , Aymeric Histace (5) , Edouard Klein (6)

1 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes
2 MIDI - Multimedia Indexation and Data Integration
3 MLIA - Machine Learning and Information Access
4 ENSEA - Ecole Nationale Supérieure de l'Electronique et de ses Applications
5 ASTRE [Cergy-Pontoise]
6 PJGN - Pôle judiciaire de la gendarmerie nationale

Most image retrieval methods use global features that aggregate local distinctive patterns into a single representation. However, the aggregation process destroys the relative spatial information by considering orderless sets of local descriptors. We propose to integrate relative spatial information into the aggregation process by taking into account co-occurrences of local patterns in a tensor framework. The resulting signature called Improved Spatial Tensor Aggregation (ISTA) is able to reach state of the art performances on well known datasets such as Holidays, Oxford5k and Paris6k.

Mots clés: Image retrieval

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Apprentissage [cs.LG]

Pierre Jacob, David Picard, Aymeric Histace, Edouard Klein. Leveraging Implicit Spatial Information In Global Features For Image Retrieval . IEEE International Conference in Image Processing, Oct 2018, Athens, Greece. ⟨hal-01816918⟩