Assoc. Prof. Kezhi Mao
Nanyang Technological University, Singapore
Mao Kezhi obtained his BEng, MEng and PhD from Jinan University, Northeastern University, and University of Sheffield in 1989, 1992 and 1998 respectively. He is now an Associate Professor at School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. His research covers a couple of subfields of artificial intelligence (AI), including machine learning, computer vision, natural language processing, and information fusion. Over the past 25 years, he has developed novel algorithms and frameworks to address various issues in the field of artificial intelligence. He has published over hundred research papers on top journals and conferences, which have received 10000+ citations (Google Scholar).
As a strong advocate of translational research, he has collaborated with government agencies and hospitals and developed a couple of prototypes of AI systems for image processing and natural language processing. He served as consultant for a number of companies such as Deloitte & Touche, ST Engineering, Zhuyi Technologies, and Rakuten Group etc, advising on R&D of AI and machine learning.
He now serves as Member of Editorial Board of Neural Networks, Academic Editor of Computational Intelligence and Neuroscience, and General Chair, General Co-Chair, Invited Panelist, and Invited Speaker of a number of international conferences.
Speech Title "Depth Estimation from Single Images: Techniques, Challenges, and Applications"
Abstract: Depth estimation is a fundamental problem in computer vision with numerous applications, including robotics and navigation, augmented reality and medical imaging etc. Traditionally, depth estimation relies on stereo vision or depth sensors, which can be expensive or impractical in many scenarios. Recent advances in deep learning and computer vision have significantly improved the accuracy and feasibility of monocular depth estimation. This talk will provide a comprehensive overview of the latest techniques, challenges, and applications of depth estimation from single images. In addition, the current limitations and future directions will be discussed.