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 over tting 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 classi er is shown to pro-
vide excellent ne-grained classi cation performances out-
performing the results reported in the literature on several
classi cation tasks.
Type de document :
Communication dans un congrès
International Conference on Multimedia Retrieval 2015, Jun 2015, Shangai, China
https://hal.inria.fr/hal-01145988
Contributeur : Alexis Joly
<>
Soumis le : lundi 27 avril 2015 - 14:19:42
Dernière modification le : jeudi 24 mai 2018 - 15:59:21
