Sabrina Caldwell

Sabrina Caldwell serves as a Senior Lecturer and Researcher at The Australian National University's School of Computing. As Co-chair of JPEG Trust within the ISO/IEC JTC 1/SC 29/WG 1 working group, she has led the development of international standard ISO/IEC 21617-1 for media trustworthiness assessment, scheduled for publication in October 2024. Her work bridges academic research with global standardization efforts in digital media security.

Her research interests focus on two interconnected domains: digital image authentication (particularly distinguishing authentic photographs from AI-manipulated content) and physiological signal processing using neural networks. Within human-AI teaming, she investigates deception detection through multimodal analysis of nonverbal behavior and develops frameworks for assessing the credibility of visual media in the age of generative AI. Her fingerprint reveals strong concentrations in neural networks (100%), physiological signal processing (85%), and affective computing (30%).

Analysis of her 29 research outputs shows increasing emphasis on multimodal deception detection and media trustworthiness standards since 2020. Her work demonstrates consistent collaboration with international researchers like Temmermans, Bhowmik, and Kyburz, with notable projects including MEC23: Advanced Biometrics and C2 Sociotechnical Experimentation. The publications reveal a strategic shift toward practical standardization frameworks after 2022, coinciding with her JPEG Trust leadership role.

Caldwell holds active research supervision registration and leads projects including the 2023-2026 Australian government travel grant for international standards participation and the 2024 Human-machine aesthetics seed grant. Her projects consistently address real-world applications of computing for social good, particularly in media trust and human-centered biometrics.