Media Forensics in the Era of Deepfakes by Dr. Scott McCloskey, Director of Computational Imaging, Kitware, Inc.
Image forensics is a longstanding area of computer vision research, in which one attempts to detect, localize, and/or attribute manipulations of still and video imagery. This talk describes the progression of the field from the foundations of digital image forensics to modern issues related to deepfakes. Early work focused on manipulations made with interactive image editing, using signal processing approaches that estimate and match sensor fingerprints. Then, in the late 2010’s, new forensic challenges arose from the rise of generative AI-based image synthesis and editing, along with the increased use of in-camera image enhancement. These rapid changes in nature of both real and altered imagery necessitated the development of new algorithms and datasets. This talk will present technical approaches from these different eras of digital image forensics, and conclude with an enumeration of current challenges in image forensics.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
Loading virtual attendance info...
- Contact Event Hosts
-
For more information contact: Hong Zhao (201)-692-2350, zhao@fdu.edu; Alfredo Tan, tan@fdu.edu; or Kalyan Mondal, mondal@fdu.edu.
- Co-sponsored by Gildart Haase School of Engineering, Fairleigh Dickinson University
- Starts 07 March 2026 01:20 AM UTC
- Ends 30 April 2026 12:22 AM UTC
- No Admission Charge
Speakers
Scott of Kitware
Media Forensics in the Era of Deepfakes
Image forensics is a longstanding area of computer vision research, in which one attempts to detect, localize, and/or attribute manipulations of still and video imagery. This talk describes the progression of the field from the foundations of digital image forensics to modern issues related to deepfakes. Early work focused on manipulations made with interactive image editing, using signal processing approaches that estimate and match sensor fingerprints. Then, in the late 2010’s, new forensic challenges arose from the rise of generative AI-based image synthesis and editing, along with the increased use of in-camera image enhancement. These rapid changes in nature of both real and altered imagery necessitated the development of new algorithms and datasets. This talk will present technical approaches from these different eras of digital image forensics, and conclude with an enumeration of current challenges in image forensics.
Biography:
Dr. Scott McCloskey is the Director of Computational Imaging at Kitware, where he leads a group of researchers and developers within the Computer Vision team to develop new capabilities involving non-traditional sensing. During his time at Kitware Dr. McCloskey has led several large, multi-organization teams for research projects funded by the Defense Advanced Research Projects Agency (DARPA), Intelligence Advanced Research Projects Activity (IARPA), Air Force Research Laboratory (AFRL), National Geospatial-Intelligence Agency (NGA), and Missile Defense Agency (MDA). In addition to leading the Computational Imaging sub-group, Dr. McCloskey founded Kitware’s Office in Minneapolis.
Dr. McCloskey remains actively involved with technical conferences in the computer vision and computational imaging communities. He has served as general chair, program chair, and area chair for the IEEE Winter Conference on the Applications of Computer Vision (WACV); as the industrial relations chair, corporate relations chair, and area chair for the premier computer vision conferences including the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) and International Conference on Computer Vision (ICCV). He has served as a reviewer for many other conferences and journals, on a National Science Foundation review panel, and as an associate editor of the Pattern Recognition journal.
Dr. McCloskey received his Ph.D. in Computer Science from McGill University in 2008, his M.S. in computer science from the Rochester Institute of Technology in 2002, and his B.S. in computer science and mathematics from the University of Wisconsin – Madison in 1998. Between his undergraduate and Ph.D., Scott completed the Image Science Career Development program at the Eastman Kodak research labs.
Email:
Address:United States