On the relevance of sparsity for image classification

R. Rigamonti, V. Lepetit, G. González, E. Türetken, F. Benmansour, M. Brown, P. Fua

Research output: Contribution to journalArticlepeer-review

9 Citations (SciVal)

Abstract

In this paper we empirically analyze the importance of sparsifying representations for classification purposes. We focus on those obtained by convolving images with linear filters, which can be either hand designed or learned, and perform extensive experiments on two important Computer Vision problems, image categorization and pixel classification. To this end, we adopt a simple modular architecture that encompasses many recently proposed models. The key outcome of our investigations is that enforcing sparsity constraints on features extracted in a convolutional architecture does not improve classification performance, whereas it does so when redundancy is artificially introduced. This is very relevant for practical purposes, since it implies that the expensive run-time optimization required to sparsify the representation is not always justified, and therefore that computational costs can be drastically reduced.
Original languageEnglish
Pages (from-to)115-127
Number of pages13
JournalComputer Vision and Image Understanding
Volume125
DOIs
Publication statusPublished - 1 Aug 2014

Fingerprint

Dive into the research topics of 'On the relevance of sparsity for image classification'. Together they form a unique fingerprint.

Cite this