Please use this identifier to cite or link to this item:
Type: Artigo de periódico
Title: Pooling in image representation: The visual codeword point of view
Author: Avila, S
Thome, N
Cord, M
Valle, E
Araujo, AD
Abstract: In this work, we propose Bossallova, a novel representation for content-based concept detection in images and videos, which enriches the Bag-of-Words model. Relying on the quantization of highly discriminant local descriptors by a codebook, and the aggregation of those quantized descriptors into a single pooled feature vector, the Bag-of-Words model has emerged as the most promising approach for concept detection on visual documents. Bossallova enhances that representation by keeping a histogram of distances between the descriptors found in the image and those in the codebook, preserving thus important information about the distribution of the local descriptors around each codeword. Contrarily to other approaches found in the literature, the non-parametric histogram representation is compact and simple to compute. Bossallova compares well with the state-of-the-art in several standard datasets: MIRFLICKR, ImageCLEF 2011, PASCAL VOC 2007 and 15-Scenes, even without using complex combinations of different local descriptors. It also complements well the cutting-edge Fisher Vector descriptors, showing even better results when employed in combination with them. Bossallova also shows good results in the challenging real-world application of pornography detection. (C) 2012 Elsevier Inc. All rights reserved.
Subject: Image classification
Image representation
Pattern recognition
Visual dictionary
Country: EUA
Editor: Academic Press Inc Elsevier Science
Citation: Computer Vision And Image Understanding. Academic Press Inc Elsevier Science, v. 117, n. 5, n. 453, n. 465, 2013.
Rights: fechado
Identifier DOI: 10.1016/j.cviu.2012.09.007
Date Issue: 2013
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.