Lecture Title: Theory and Technology of Multimodal Fusion for Low-Quality Data
Speaker: Prof. Zhang Changqing
Venue: Room 9322, Xipu Campus, Southwest Jiaotong University

Abstract:
Multimodal data often exhibit complex interrelationships, and in real-world scenarios, the quality of data from different information sources can vary dynamically across time, space, and samples. This dynamic nature significantly impacts the ability to support tasks effectively. For instance, multimodal data may be affected by adverse weather conditions or sensor malfunctions, where radar signals are often more reliable than RGB camera signals during nighttime or in rainy and foggy weather. Based on this context, this lecture will introduce machine learning methods and applications designed for low-quality multimodal data and analyze the theoretical foundations behind them.

Speaker Biography:
Prof. Zhang Changqing is a professor and doctoral advisor at the College of Intelligence and Computing, Tianjin University, and Vice Dean of the School of Artificial Intelligence. He is a recipient of a national-level youth talent award. His research focuses on machine learning and computer vision, with multiple papers selected for oral presentations or spotlight papers at ICML, CVPR, and NeurIPS. His work has been cited over 10,000 times on Google Scholar. Prof. Zhang’s achievements include the First Prize of the Natural Science Award from the China Society of Image and Graphics, the Best Paper Award at ICME, and recognition in Baidu's list of Global High-Potential Chinese AI Scholars, as well as Stanford University's Global Top 2% Scientists list.