Image Phylogeny by Minimal Spanning Trees

Dias, Zanoni; Rocha, Anderson; Goldenstein, Siome

doi:10.1109/tifs.2011.2169959

Cited by 75 publications

(106 citation statements)

References 28 publications

Supporting

Mentioning

106

Contrasting

Order By: Relevance

“…The most time consuming operation in VPT reconstruction is the estimation of T . Then, the underlying VPT is estimated by means of optimization strategies that identify the maximum/minimum spanning tree, like Oriented Kruskal (OK) [12] or Optimum Branching (OB) [13].…”

Section: Problem Statement and Related Workmentioning

confidence: 99%

Video phylogeny tree reconstruction using aging measures

Milani

Bestagini

Tubaro

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Abstract-The increasing diffusion of user-friendly editing software and online media sharing platforms has brought forth a growing on-line availability of near-duplicate (ND) videos. The need of authenticating these contents and tracing back their history has led to the investigation of forensic algorithms for the reconstruction of the video phylogeny tree (VPT), i.e., an acyclic directed graph summarizing video genealogical relationships. Unfortunately, state-of-the-art solutions for VPT reconstruction suffer from strong computational requirements.In this paper, we propose a processing age measure based on video DCT coefficients and motion vectors statistics, which enables to provide preliminary information about possible video parent-child relationship. The use of processing age allows a forensic analyst to blindly select a smaller amount of significant video pairs to be compared for VPT reconstruction. This solution grants computational complexity reduction to the overall VPT reconstruction pipeline.

show abstract

Section: Problem Statement and Related Workmentioning

confidence: 99%

Video phylogeny tree reconstruction using aging measures

Milani

Bestagini

Tubaro

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…It consists of tackling the relationship between object duplicates of one digital object (e.g., image near duplicates). In other words, given a set of similar, but not identical, copies of an image, Image Phylogeny algorithms [18,16] are able to construct a hierarchical relationship among them showing their relationship over time.…”

Section: Multimedia Phylogenymentioning

confidence: 99%

Visual Computing and Machine Learning Techniques for Digital Forensics

Carvalho

Pedrini

Rocha

2015

RITA

View full text Add to dashboard Cite

A rapidez com que diversos campos da ciência vêm avançando dia após dia é notável. Em especial, os avanços tecnológicos têm impressionado muitas pessoas, uma vez que introduzem em suas vidas fatos que vão além da imaginação. Inspirados em métodos anteriormente apresentados por programas de ficção científica, a comunidade ligada à área de Computação criou um novo campo de pesquisa denominado Análise Forense de Documentos Digitais, o qual foca em desenvolver e implantar métodos capazes de auxiliar na luta contra crimes digitais, tais como a falsificação de imagens digitais. Este artigo apresenta alguns dos principais conceitos relacionados à Análise Forense de Documentos Digitais e, de forma complementar, apresenta algumas das mais recentes e poderosas técnicas baseadas em conceitos provenientes das áreas de Computação Gráfica, Processamento de Imagens, Visão Computacional e Aprendizado de Máquina para detectar falsificações em fotografias digitais. Alguns tópicos que são abordados neste trabalho incluem: atribuição de fonte, detecção de ataques, detecção de pornografia, filogenia multimídia e detecção de falsificações. Por fim, este trabalho destaca os desafios e problemas em aberto no campo da Análise Forense de Imagens Digitais para que os leitores se familiarizem com as oportunidades de pesquisa disponíveis. Visual Computing and Machine Learning Techniques for Digital ForensicsAbstract: It is impressive how fast science has improved day by day in so many different fields. In special, technology advances are shocking so many people bringing to their reality facts that previously were beyond their imagination. Inspired by methods earlier presented in scientific fiction shows, the computer science community has created a new research area named Digital Forensics, which aims at developing and deploying methods for fighting against digital crimes such as digital image forgery. This work presents some of the main concepts associated with Digital Forensics and, complementarily, presents some recent and powerful techniques relying on Computer Graphics, Image Processing, Computer Vision and Machine Learning concepts for detecting forgeries in photographs. Some topics addressed in this work include: source attribution, spoofing detection, pornography detection, multimedia phylogeny, and forgery detection. Finally, this work highlights the challenges and open problems in Digital Image Forensics to provide the readers with the myriad opportunities available for research.

show abstract

“…Although most of the changes related to near-duplicate multimedia objects are natural and not necessarily harmful, sometimes the distribution itself might cause copyright infringement or even represent a criminal action [4,5]. In some situations, the spreading pattern of an image or video can help companies to understand demographics and effectiveness of an ad campaign or a product.…”

Section: Introductionmentioning

confidence: 99%

“…These scenarios motivated the dawning of a new research subfield called Multimedia Phylogeny [4], to investigate the history and evolutionary process of digital objects. In other words, we are looking for the structure of modifications of multimedia objects.…”

Section: Introductionmentioning

confidence: 99%

“…Only recently there have been the first attempts to go beyond the near-duplicate identification problem to pinpoint the structure of relationships within a set of objects [9,4,10,8,7]. However, these early investigations are constrained to the case of image and video near-duplicates, which are related by a set of possible transformations -e.g., cropping, affine warping (considering as special cases resampling, rotation and translation), brightness/contrast adjustment and lossy compression.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation