Finding Patterns in Multimodal Data

Abstract: 
This talk explores the expanding role of computer vision and machine learning in detecting meaningful patterns across diverse multimodal datasets. Moving beyond traditional object recognition, we will examine how these methods are applied to uncover abnormalities in complex spatio-temporal data - ranging from brain connectomes to movement patterns in ecological and security contexts. We will also discuss how imaging and non-imaging data are being reimagined to tackle challenges such as malware detection and classification. The talk will conclude with an introduction to the BisQue platform designed to promote reproducibility in computer vision and facilitate interdisciplinary collaboration by enabling the sharing of methods and data across domains.

Speaker bio:
B.S. Manjunath holds a B.E. in electronics engineering from Bangalore University, M.E. in systems science and automation from the Indian Institute of Science, and a Ph.D. in electrical engineering from the University of Southern California. He is currently a Distinguished Professor of  Electrical and Computer Engineering at the University of California, Santa Barbara, with an affiliated appointment in Computer Science. His research focuses on computer vision and machine learning, with extensive applications across various scientific fields. Manjunath has been instrumental in developing BisQue, a robust, open-source cloud-based platform designed for reproducible computer vision. He has supervised over 50 PhD theses, authored more than 350 peer-reviewed publications, and holds 25 US patents. His contributions to the field have been recognized with the 2020 IEEE CS Edward J. McCluskey Technical Achievement Award.
He is a fellow of IEEE, ACM, AIMBE, and the National Academy of Inventors (NAI).