The rapid advance of artificial intelligence (AI) has led to several emerging technologies, such as Generative Adversarial Networks (GANs), deepfakes, CGI and anti-forensics techniques, which significantly threaten the trustworthiness of media content. To detect inadvertent misinformation or deliberate deception via disinformation and ensure digital content trust and authentication, we provide public researchers with a comprehensive evaluation platform, the Open Media Forensic Challenge (OpenMFC) (https://mfc.nist.gov), to develop media forensic technologies to automatically detect unauthentic imagery (i.e., images and videos) and retrieve the digital content provenance.
The OpenMFC is an online evaluation series open to public participants worldwide to support and promote media forensics research and help advance the state-of-the-art technologies. It is a continuation of the NIST Media Forensics Challenge’s (MFC) efforts in supporting the DARPA MediFor Program (2017-2020).
In this talk, we will give an overview of the OpenMFC program, including the program design, major challenges, evaluation tasks, released datasets, evaluation reports from past years, and the future evaluation plan.
Keywords: Open Media Forensic Challenge (OpenMFC), Generative Adversarial Networks (GANs), Deepfakes, Artificial Intelligence (AI) Algorithm Evaluation
Dr. Haiying Guan received her PhD degree in Computer Science from the University of California Santa Barbara, United States. She is a Computer Scientist in the Multimedia Information Group (MIG), Information Access Division (IAD), Information Technology Laboratory (ITL) at National Institute of Standards and Technology (NIST). She is currently the technical lead of Open Media Forensics Challenge (OpenMFC) evaluation program. She was a co-PI on Media Forensics Challenge (MFC) evaluation project sponsored by DARPA (Defense Advanced Research Projects Agency) MediFor program. She has government, industry, and academic research experiences in computer vision, media forensics, deepfakes, biometrics, medical image processing, video analytics and quality measures, and human-computer interaction. Her current research interests focus on the evaluation program design and benchmark dataset collection for Artificial Intelligence (AI) system evaluation.