Local Convolutional Features with Unsupervised Training for Image Retrieval

Mattis Paulin 1 Matthijs Douze 1 Zaid Harchaoui 1 Julien Mairal 1 Florent Perronnin 2 Cordelia Schmid 1
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 Xerox Research Centre Europe [Meylan]
Abstract : Patch-level descriptors underlie several important computer vision tasks, such as stereo-matching or content-based image retrieval. We introduce a deep convolutional architecture that yields patch-level descriptors, as an alternative to the popular SIFT descriptor for image retrieval. The proposed family of descriptors, called Patch-CKN, adapt the recently introduced Convolutional Kernel Network (CKN), an unsupervised framework to learn convolutional architectures. We present a comparison framework to benchmark current deep convolutional approaches along with Patch-CKN for both patch and image retrieval, including our novel ``RomePatches'' dataset. Patch-CKN descriptors yield competitive results compared to supervised CNNs alternatives on patch and image retrieval.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [42 references]  Display  Hide  Download

figure_deep_kernel_2layer.jpg
figure_deep_kernel_2layer.jpg
https://hal.inria.fr/hal-01207966
Contributor : Thoth Team <>
Submitted on : Thursday, October 1, 2015 - 4:27:12 PM
Last modification on : Tuesday, February 12, 2019 - 10:30:06 AM
Document(s) archivé(s) le : Saturday, January 2, 2016 - 11:20:59 AM

Files

deep_patches.pdf
Files produced by the author(s)

Identifiers

Collections

INRIA | LJK | UGA | LJK_GI | LJK_GI_LEAR

Citation

Mattis Paulin, Matthijs Douze, Zaid Harchaoui, Julien Mairal, Florent Perronnin, et al.. Local Convolutional Features with Unsupervised Training for Image Retrieval. ICCV - IEEE International Conference on Computer Vision, Dec 2015, Santiago, Chile. pp.91-99, ⟨10.1109/ICCV.2015.19⟩. ⟨hal-01207966⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

1647

Files downloads

12834

Contact


archive-ouverte@inria.fr