Discriminative learning of local image descriptors

Matthew A Brown, Gang Hua, S Winder

Research output: Contribution to journalArticle

296 Citations (Scopus)
334 Downloads (Pure)

Abstract

In this paper, we explore methods for learning local image descriptors from training data. We describe a set of building blocks for constructing descriptors which can be combined together and jointly optimized so as to minimize the error of a nearest-neighbor classifier. We consider both linear and nonlinear transforms with dimensionality reduction, and make use of discriminant learning techniques such as Linear Discriminant Analysis (LDA) and Powell minimization to solve for the parameters. Using these techniques, we obtain descriptors that exceed state-of-the-art performance with low dimensionality. In addition to new experiments and recommendations for descriptor learning, we are also making available a new and realistic ground truth data set based on multiview stereo data.
Original languageEnglish
Pages (from-to)43-57
Number of pages15
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume33
Issue number1
Early online date25 Feb 2010
DOIs
Publication statusPublished - Jan 2011

Fingerprint Dive into the research topics of 'Discriminative learning of local image descriptors'. Together they form a unique fingerprint.

  • Cite this