Coming as a great result from a partnership between RECOD and two Italian universities (University of Florence and University of Siena) a paper on Multiple Parenting Phylogeny was accepted on IEEE Transaction on Information Security and Forensics. The work was supported in part by CAPES through the DeepEyes Project, CNPq, FAPESP and, finally, by the European Union through the Reverse Engineering of Audio-Visual Content Data Project.
The paper, entitled Multiple Parenting Phylogeny Relationships in Digital Images, advances the start of the art of finding all the parenting relationships existent in a set of Near-Duplicate Images. To do so, the authors take into consideration cases when one image is created through the combination of parts from different images. The goal is to discover the different phylogenies and to establish multiple parenting relationships in a set of images with varied contents. The impact of the proposed solution lies down on practical scenarios such as content tracking, forensics and copyright enforcement. Here’s the abstract:
Recently, several works have been concerned with modeling the parenthood relationships between near duplicates in a set of images. Two images share a parenthood relationship if one is obtained by applying transformations to the other. However, this is not the only form of parenting that can exist among images. An image might be a composition created through the combination of the semantic information existent in two or more source images, establishing a relationship between the sources and the composite. The problem of identifying these relations in a set containing near-duplicate subsets of source and composition images is referred to as Multiple Parenting Phylogeny. Thus far, researchers tackled this problem with a three-step solution: (1) separation of near-duplicate groups; (2) classification of the relations between the groups; and (3) identification of the images used to create the original composition. In this work, we extend upon this framework by introducing key improvements, such as better identification of when two images share content, and improved ways to compare this content. In addition, we also introduce a new realistic professionally-created dataset of com- positions involving multiple parenting relationships. The method we present herein is properly evaluated through quantitative metrics, established for assessing the accuracy in finding multiple parenting relationships. Finally, we discuss some particularities of the framework, such as the importance of an accurate reconstruction of phylogenies and the method’s behavior when dealing with more complex compositions.
Alberto Oliveira, Pasquale Ferrara, Alessia De Rosa, Alessandro Piva, Mauro Barni, Siome Goldenstein, Zanoni Dias, Anderson Rocha. Multiple Parenting Phylogeny Relationships in Digital Images. IEEE Transaction on Information Security and Forensics, 2015.