Prof. Adrian Munteanu
Vrjie Universiteit Brussel, Belgium
Title:
AI-Based Multimodal Sensor Processing
Bio:
Adrian Munteanu is a Full Professor at the Electronics and Informatics (ETRO) Department of the Vrije Universiteit Brussel (VUB), Belgium. He received the MSc degree in Electronics and Telecommunications from Politehnica University of Bucharest, Romania, in 1994, the MSc degree in Biomedical Engineering from University of Patras, Greece, in 1996, and the Doctorate degree in Applied Sciences (Summa Cum Laude) from Vrije Universiteit Brussel, Belgium, in 2003. From 2004 to 2010, he was a post-doctoral fellow with the Fund for Scientific Research – Flanders (FWO), Belgium, and since 2007, he is a Professor at VUB.
His research interests include image, video and 3D graphics compression, error-resilient coding, multimedia transmission over networks, 3D video, distributed visual processing, and AI-driven signal processing and sensing systems. Adrian Munteanu is the author of more than 400 journal and conference publications, book chapters, and contributions to standards, and holds several patents in image and video coding. He is the recipient of the 2004 BARCO-FWO prize for his PhD work, a (co-)recipient of fifteen other international scientific prizes and awards. Adrian Munteanu served as Associate Editor for IEEE Signal Processing Letters, IEEE Transactions on Multimedia, and IEEE Transactions on Image Processing.
Abstract:
This keynote presents a historical overview of AI-driven robust multimodal sensor processing developed in the speaker’s lab. It focuses on lightweight AI tools for low-level tasks such as sensor denoising, calibration, and cross-modal alignment, emphasizing reliability under real-world noise and drift, and feasibility on resource-constrained platforms.
The talk then illustrates how these methods extend to diverse applications, including anthropometric measurement extraction, multimodal video processing, radiance field representations, and Hall sensing. The keynote closes with a forward-looking discussion on scalable, trustworthy multimodal sensing pipelines and unified representations linking signal processing and perception.