Physiological forensics Chau-Wai Wong (North Carolina State University, United States) and Min Wu (University of Maryland – College Park, United States) – Virtual presentation
Physiological signals have been playing important roles in our life. This tutorial will provide an overview of the recent development of physiological forensics. We shall focus on physiological sensing from non-conventional and opportunistic modalities, such as heartbeat signal from video without requiring contact, and the respective applications as well as security and privacy issues.In the first part of the tutorial, we plan to introduce various types of signals and the methodologies of acquisition, extraction, and postprocessing in the context of physiologic forensics. We will examine through the lens of “micro signals” in terms of how cutting-edge signal processing and learning algorithms can deal with extremely low signal fidelity scenarios. In the second part of the tutorial, we plan to introduce physiological applications in health monitoring, surveillance, media forensics and discuss security issues under adversaries and privacy implications. For example, we will examine how the blood volume signal can be exploited for deepfake detection, how it may be forged to circumvent the attribution of an individual, and how privacy concerns in data collection may be alleviated. We will also identify opportunities for the Information Forensics and Security (IFS) community to contribute to the cross-disciplinary research of physiological sensing and applications.