HAL-Inria

Publications, software ... of Inria's scientists

Kernelizing Spatially Consistent Visual Matches for Fine-Grained Classification

Valentin Leveau ^{1, 2, 3} Alexis Joly ^{2, 3} Olivier Buisson ⁴ Patrick Valduriez ^{5, 3, 6}

1 INA - Institut National de l'Audiovisuel

2 LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier

3 ZENITH - Scientific Data Management

LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée

4 INA-SUP

5 MSR - INRIA - Microsoft Research - Inria Joint Centre

6 IBC - Institut de Biologie Computationnelle

Abstract : This paper introduces a new image representation relying on the spatial pooling of geometrically consistent visual matches. We therefore introduce a new match kernel based on the in- verse rank of the shared nearest neighbors combined with local geometric constraints. To avoid overtting and reduce processing costs, the dimensionality of the resulting over- complete representation is further reduced by hierarchically pooling the raw consistent matches according to their spa- tial position in the training images. The nal image repre- sentation is obtained by concatenating the resulting feature vectors at several resolutions. Learning from these represen- tations using a logistic regression classier is shown to pro- vide excellent ne-grained classication performances out- performing the results reported in the literature on several classication tasks.

Document type :

Conference papers

Domain :

Computer Science [cs] / Computer Vision and Pattern Recognition [cs.CV]

Computer Science [cs] / Data Structures and Algorithms [cs.DS]

Computer Science [cs] / Multimedia [cs.MM]

Complete list of metadatas

https://hal.inria.fr/hal-01145988
Contributor : Alexis Joly <>
Submitted on : Monday, April 27, 2015 - 2:19:42 PM
Last modification on : Thursday, October 10, 2019 - 11:46:03 AM