Toward Practical Object Trackers: From Feature Combination to Modality Fusion

Abstract

Visual object tracking is an important and active research topic in computer vision community because of its wide range of applications, e.g., intelligent video surveillance and human computer interaction. Although visual object tracking has been extensively studied in the last decade, it still remains challenging to develop a robust tracking algorithm due to appearance variations and background distractors in practical scenarios. In this talk, I will present our research work in visual object tracking. First, I will introduce two feature fusion models for object tracking in RGB videos, which dynamically select reliable features and exploit consistency and complementarity of the fused features to handle the variations of the tracked objects. Then a robust collaborative discriminative learning model, which integrates information from heterogeneous RGB and infrared modalities for tracking, will also be illustrated. Finally, this talk will conclude with a brief introduction and discussion of our ongoing work and potential future research of tracking in other fields (e.g. video processing in medical applications).

Biography

Dr. Lan is currently a Post-doctoral Research Fellow at the Department of Computer Science, Hong Kong Baptist University. He was a visiting scholar at the Center for Automation Research, UMIACS, University of Maryland at College Park from January to July in 2015 and a visiting researcher at the Vision and Learning Lab in the University of California at Merced in February of 2017. He received his Ph.D. degree from the Department of Computer Science, Hong Kong Baptist University in 2016, and his B. Eng. degree in Computer Science and Technology from South China University of Technology in 2012. He was awarded the IBM China Excellent Student Scholarship from the China Scholarship Council in 2012. His research interests include artificial intelligence, computer vision and pattern recognition. He has served as invited reviewers for journals including TIP, TNNLS, TCYB, TIFS, and program committee members for conferences including AAAI, IJCAI, ECAI.