Triangulation embedding and democratic aggregation for image search

Conference item

We consider the design of a single vector representation for an image that embeds and aggregates a set of local patch descriptors such as SIFT. More specifically we aim to construct a dense representation, like the Fisher Vector or VLAD, though of small or intermediate size. We make two contributions, both aimed at regularizing the individual contributions of the local descriptors in the final representation. The first is a novel embedding method that avoids the dependency on absolute distances by encoding directions. The second contribution is a "democratization" strategy that further limits the interaction of unrelated descriptors in the aggregation stage. These methods are complementary and give a substantial performance boost over the state of the art in image search with short or mid-size vectors, as demonstrated by our experiments on standard public image retrieval benchmarks.

Authors: Jégou, H; Zisserman, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: 2014 IEEE Conference on Computer Vision and Pattern Recognition
• Pages: 3310-3317
• Publication date: 2014-09-25
• Acceptance date: 2014-03-01
• Event title: 27th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2014)
• Event location: Columbus, Ohio, USA
• Event website: https://www.computer.org/csdl/proceedings/cvpr/2014/12OmNCbCrVT
• Event start date: 2014-06-23
• Event end date: 2014-06-28
• DOI: 10.1109/cvpr.2014.417
• EISSN: 1063-6919
• ISSN: 1063-6919
• EISBN: 9781479951185
• Language: English
• Keywords: eigenvalues and eigenfunctions; convergence; encoding; vectors; equations; optimization; kernel