Convolutional neural network architecture for geometric matching - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Pattern Analysis and Machine Intelligence Année : 2018

Convolutional neural network architecture for geometric matching

Résumé

We address the problem of determining correspondences between two images in agreement with a geometric model such as an affine, homography or thin-plate spline transformation, and estimating its parameters. The contributions of this work are threefold. First, we propose a convolutional neural network architecture for geometric matching. The architecture is based on three main components that mimic the standard steps of feature extraction, matching and simultaneous inlier detection and model parameter estimation, while being trainable end-to-end. Second, we demonstrate that the network parameters can be trained from synthetically generated imagery without the need for manual annotation and that our matching layer significantly increases generalization capabilities to never seen before images. Finally, we show that the same model can perform both instance-level and category-level matching giving state-of-the-art results on the challenging PF, TSS and Caltech-101 datasets.
Fichier principal
Vignette du fichier
cnngeometric_pami.pdf (21.37 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01859616 , version 1 (22-08-2018)

Identifiants

Citer

Ignacio Rocco, Relja Arandjelovic, Josef Sivic. Convolutional neural network architecture for geometric matching. IEEE Transactions on Pattern Analysis and Machine Intelligence, In press, pp.1-14. ⟨10.1109/TPAMI.2018.2865351⟩. ⟨hal-01859616⟩
304 Consultations
859 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More