U-Phylogeny: Undirected provenance graph construction in the wild

Bharati, Aparna; Moreira, Daniel; Pinto, Allan; Brogan, Joel; Bowyer, Kevin W.; Flynn, Patrick J.; Scheirer, Walter J.; Rocha, Anderson

doi:10.1109/icip.2017.8296535

Cited by 16 publications

(22 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Provenance analysis methods [8] have been used to identify the donor images of a composite. In such cases, undirected phylogeny trees [9] are suitable, since deducing the links is more critical than determining the direction of the link. Image phylogeny trees (IPTs), on the other hand, deal with the task of determining ancestral relationships, i.e., explicitly determining the parent and child node as well as the predecessors.…”

Section: Related Workmentioning

confidence: 99%

Face Phylogeny Tree Using Basis Functions

Banerjee

Ross

2020

IEEE Trans. Biom. Behav. Identity Sci.

View full text Add to dashboard Cite

Photometric transformations, such as brightness and contrast adjustment, can be applied to a face image repeatedly creating a set of near-duplicate images. Identifying the original image from a set of such near-duplicates and deducing the relationship between them are important in the context of digital image forensics. This is commonly done by generating an image phylogeny treea hierarchical structure depicting the relationship between a set of near-duplicate images. In this work, we utilize three different families of basis functions to model pairwise relationships between near-duplicate images. The basis functions used in this work are orthogonal polynomials, wavelet basis functions and radial basis functions. We perform extensive experiments to assess the performance of the proposed method across three different modalities, namely, face, fingerprint and iris images; across different image phylogeny tree configurations; and across different types of photometric transformations. We also utilize the same basis functions to model geometric transformations and deep-learning based transformations. We also perform extensive analysis of each basis function with respect to its ability to model arbitrary transformations and to distinguish between the original and the transformed images. Finally, we utilize the concept of approximate von Neumann graph entropy to explain the success and failure cases of the proposed IPT generation algorithm. Experiments indicate that the proposed algorithm generalizes well across different scenarios thereby suggesting the merits of using basis functions to model the relationship between photometrically and geometrically modified images. !• Both the authors are with the

show abstract

Section: Related Workmentioning

confidence: 99%

Face Phylogeny Tree Using Basis Functions

Banerjee

Ross

2020

IEEE Trans. Biom. Behav. Identity Sci.

View full text Add to dashboard Cite

show abstract

“…As a step towards a more general framework for image provenance analysis inspired by image phylogeny works, recent work on undirected provenance graph construction [11] adopted a more general taxonomy and dataset proposed by the American National Institute of Standards and Technology (NIST) [55]. It offered the U-phylogeny pipeline as a preliminary approach towards solving provenance analysis, which is not restricted to either a closed set of image transformations, or the number of donor images to form multi-parent composites.…”

Section: Related Workmentioning

confidence: 99%

“…Once the images are described, for each image pair, the p most relevant interest points of each image are matched using brute-force pairwise comparison based on the L2 distance between the descriptors. The best matched correspondences are filtered to retain the geometrically consistent ones, as described in [11]. As a consequence, a symmetric adjacency matrix is obtained with the quantity of matched interest points between each pair of images.…”

Section: Adjacency Matrix Computationmentioning

confidence: 99%

Beyond Pixels: Image Provenance Analysis Leveraging Metadata

Bharati

Moreira

Brogan

et al. 2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

Creative works, whether paintings or memes, follow unique journeys that result in their final form. Understanding these journeys, a process known as "provenance analysis," provides rich insights into the use, motivation, and authenticity underlying any given work. The application of this type of study to the expanse of unregulated content on the Internet is what we consider in this paper. Provenance analysis provides a snapshot of the chronology and validity of content as it is uploaded, re-uploaded, and modified over time. Although still in its infancy, automated provenance analysis for online multimedia is already being applied to different types of content. Most current works seek to build provenance graphs based on the shared content between images or videos. This can be a computationally expensive task, especially when considering the vast influx of content that the Internet sees every day. Utilizing non-contentbased information, such as timestamps, geotags, and camera IDs can help provide important insights into the path a particular image or video has traveled during its time on the Internet without large computational overhead. This paper 1 tests the scope and applicability of metadata-based inferences for provenance graph construction in two different scenarios: digital image forensics and cultural analytics.

show abstract

“…Recently, researchers began to actively study IPA, and a few approaches are introduced in [5,[8][9][10][11]. But the IPA systems still have much room to improve their performance.…”

Section: Introductionmentioning

confidence: 99%

Media Forensics Challenge image provenance evaluation and state-of-the-art analysis on large-scale benchmark datasets

Jin¹,

Lee²,

Fiscus³

et al. 2020

View full text Add to dashboard Cite

With the development of storage, transmission, editing, and sharing tools, digital forgery images are propagating rapidly. The need for image provenance analysis has never been more timely. Typical applications are content tracking, copyright enforcement, and forensics reasoning. However, large-scale image provenance datasets, which contain diverse manipulation history graphs with various manipulation operations and rich metadata, are still needed to facilitate the research. It is one of the major factors that hinders the development of techniques for image provenance analysis. To address this issue, we introduce large-scale benchmark datasets for provenance analysis, namely Media Forensics Challenge-Provenance (MFC-Prov) datasets. Two provenance tasks are designed along with evaluation metrics. Furthermore, extensive analysis is conducted for system performance in terms of accuracy on our datasets.

show abstract

U-Phylogeny: Undirected provenance graph construction in the wild

Cited by 16 publications

References 19 publications

Face Phylogeny Tree Using Basis Functions

Face Phylogeny Tree Using Basis Functions

Beyond Pixels: Image Provenance Analysis Leveraging Metadata

Media Forensics Challenge image provenance evaluation and state-of-the-art analysis on large-scale benchmark datasets

Contact Info

Product

Resources

About