Media Forensics Integrity Analytics
Abstract
The goal of this research was to develop a set of forensics tools to determine the integrity, semantic consistency and evolutionary history of images and videos. We used a data-driven approach. In TA1.1 we designed machine-learning methods trained to analyze the integrity of images and videos. In TA1.2 we used the physical integrity of the scene and the traces of electrical network frequency (ENF) to determine the location and integrity of a video. Our work in TA1.3 generated techniques to correlate the existing objects in the pool in space (spatial coherence) and time (time coherence). We have participated in the NIST evaluations and have delivered our software tools via the APIs for integration with the TA2 efforts.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2022
- Accession Number
- AD1179160
Entities
People
- C.-c. J. Kuo
- Edward J. Delp
- Luisa A. Verdolvia
- Mauro Barni
- Nasir Memon
- Stefano Tubaro
- Wael AbdAlmageed
- Walter J. Scheirer
Organizations
- Purdue University