Progressive randomization: Seeing the unseen

Autor(es)

Rocha, Anderson ^[1] ; Goldenstein, Siome ^[1]

Afiliação do(s) autor(es)

^[1] Univ Estadual Campinas, Inst Comp, BR-13084851 Campinas, SP - Brazil

• Tipo de documento: Artigo Científico
• Fonte: COMPUTER VISION AND IMAGE UNDERSTANDING; v. 114, n. 3, p. 349-362, MAR 2010.
• Citações Web of Science: 7

Resumo

In this paper, we introduce the progressive randomization (PR): a new image meta-description approach suitable for different image inference applications such as broad class Image Categorization Forensics and, Steganalysis. The main difference among PR and the state-of-the-art algorithms is that it is based on progressive perturbations on pixel values of images. With such perturbations, PR captures the image class separability allowing us to successfully infer high-level information about images. Even when only a limited number of training examples are available, the method still achieves good separability, and its accuracy increases with the size of the training set. We validate the method using two different inference scenarios and four image databases. (C) 2009 Elsevier Inc. All rights reserved. (AU)

• Processo FAPESP: 07/52015-0 - Métodos de aproximação para computação visual
• Beneficiário: Jorge Stolfi
• Modalidade de apoio: Auxílio à Pesquisa - Temático
• Processo FAPESP: 08/08681-9 - Análise Forense de Imagens: Detecção de falsificações digitais e spoofing
• Beneficiário: Anderson de Rezende Rocha
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado
• Processo FAPESP: 05/58103-3 - Classificadores e aprendizado em processamento de imagens e visão computacional
• Beneficiário: Anderson de Rezende Rocha
• Modalidade de apoio: Bolsas no Brasil - Doutorado