Dr. Lim Kian Ming
Academic Coordinator for Google Web Academy and Amazon Web Services,
Faculty of Information Science and Technology, Multimedia University, Malaysia.
Dr. Lim Kian Ming received B.IT (Hons) in Information Systems Engineering, Master of Engineering Science (MEngSc) and Ph.D. (I.T.) degrees from Multimedia University. His Ph.D. dissertation emphasizes on deep neural networks for pattern recognition. He has published a number of influential publications on Machine Learning, Computer Vision, and Pattern Recognition. His research works has also qualified him for several innovation awards at the University, National and International levels. He has filed some copyrights for his innovations. In addition, he is also active in academic curriculum. He is the Programme Co-coordinator of B.IT (Hons) Artificial Intelligence. In addition, he is also the Academia Coordinator for Google Web Academy and Amazon Web Services. He is currently a Lecturer with the Faculty of Information Science and Technology, Multimedia University. His research interests include machine learning, deep learning, computer vision, and pattern recognition.
Tutorial Title: Deep Learning and Its Applications in Video Analytics.
Deep learning has recently achieved very promising results in a wide range of areas such as computer vision, natural language processing and speech recognition. In contrast to hand-crafted methods, deep learning aims to learn hierarchical representations from large-scale data (e.g. images and videos) via deep architecture models with multiple layers of non-linear transformations. As compared to hand-crafted features, it is easier to achieve higher performance using the learned hierarchical representations. The principle behind the success of deep learning is that it is able to extract different levels of abstractions embedded in the data by carefully design the layer’s depth and width. Then, the features that are beneficial for the learning tasks are properly selected.
In recent years, video analytics has attracted increasing interests from both academic and industry. The main goal of video analytics is to automatically recognize the temporal and spatial events in videos. Thanks to the enormous advancements achieved in deep learning, recent improvements in video analytics, ranging from the applications of object tracking, object detection, human-computer interaction, and video surveillance; has automated many tasks in the industries.
This tutorial commences with the concept of deep learning and followed by its applications in video analytics.