CAP Seminar Series
Title: Solving the Fusion Problem: Model-Based Refinement and Deep Learning Synthesis
Speaker: Junfeng Wu, Chinese University of Hong Kong Shenzhen, China
Venue: EENG 909B
Date and Time: Thursday 04 December 2025, 14:00-15:00
Abstract: Effective data fusion requires solving two types of problems: refining the models we have, and creating new systhesis for problems we can't easily define. This talk delves into both through two specific research advancements. The first paper, "Supervisory Measurement-Guided Noise Covariance Estimation," addresses the refinement problem. It introduces a computationally efficient, bilevel optimization grounded in a novel likelihood factorization. This factorization converts a nested Bayesian network into a chain structure, enabling parallel State and Derivative Filters to precisely estimate sensor noise covariances and make a known model-based estimator perform optimally. The second paper, "SonarSweep," addresses the synthesis problem. In the visually degraded underwater environment, traditional geometric models break down. SonarSweep uses a deep learning pipeline to synthesize a fusion function from sonar and vision data, enabling 3D reconstruction where no explicit model exists. Together, these papers showcase the spectrum of modern data fusion, from making principled algorithms smarter to creating entirely new capabilities through learned perception.
Biography: Professor Junfeng Wu is an Associate Professor of the Chinese University of Hong Kong Shenzhen. He received his BA in the Department of Control Science and Engineering from Zhejiang University in 2009. He got his Ph.D. in the Department of Electronic and Computer Engineering in 2013. Professor Wu was a Research Associate in HKUST in 2013, and a Researcher in KTH from 2014 to 2017. From 2017 to 2021, he held a faculty position at Zhejiang University. His research interest mainly focuses on cyber security and privacy, optimization, distributed systems, control network systems, Kalman filtering, and signal processing. He was honored with a few awards, In 2019, he received Japan JSPS Invitational Fellowship. In 2014, he was honored as an outstanding reviewer by IEEE Trans. Control of Network Systems and Automatica. In 2015, he was honored Guan Zhaozhi Best Paper Award of the 34th China Control Council (CCC2015). In 2016, he received Endeavour Research Fellowship from the Australian Government.
Upcoming talks of CAP Seminar Series | Past talks of CAP Seminar Series