Tutorial Description

Recently we have witnessed an explosive growth of image/video/audio data in both the local centers and social-networking websites, such as Flickr, YouTube, and Facebook. Artificial intelligence techniques have proven useful for interpreting this preponderance of data. In the last decades, many quality models have been proposed. Computational quality models evaluate multimedia contents either objectively or subjectively, based on which large-scale multimedia content can be managed efficiently. They are useful tools in various applications such as multimedia retrieval, recommendation systems, graphical design, and etc. Building a successful quality model depends on a wide range of domain knowledge, such as multimedia, computer vision, machine learning, and even cognitive science. Extensive research efforts have been dedicated to design multimedia quality models. While effective methods to manipulate this task are still at their infancy. In detail, some key technical challenges are: 1) the deemphasized role of semantic content that are many times more important than low-level visual features in media quality prediction; 2) the necessity to incorporate human perception of multimedia contents (e.g., biologically-inspired visual/acoustic features) for quality assessment; 3) the difficulty to optimally fuse low-level and high- level visual features into a quality model; and 4) the lack of publicly available data sets to fairly evaluate the performance of a specific quality model. This tutorial targets the recent technical theory and applications on computational models, such as photo/video quality-based retargeting and feature selection for multimedia retrieval. A brief outline of our tutorial can be described as follows:

• New computational models for image and video quality evaluation (by Luming Zhang);
• Applications closely related to computational quality models, such as image cropping/retargeting, video summarization, and feature selection for efficient multimedia retrieval (by Luming Zhang);
• Discovering biologically/psychologically-inspired visual features for computational quality models (by Luming Zhang);
• Novel feature selection algorithms for multimedia analysis (by Yi Yang);
• State-of-the-art feature engineering techniques for multimedia event detection (by Yi Yang);

Biographies of Organizers

Luming Zhang received his Ph.D. degree in computer science from Zhejiang University, China. Currently he is a Postdoc Senior Research Fellow at the School of Computing, National University of Singapore. His research interests mainly include multimedia analysis, image enhancement, and pattern recognition. He has authored and co- authored more than 40 scientific articles at top venues including IEEE T-IP, T-MM, T-CYB, CVPR, and ACM MM. He served/is serving as the Guest editor for nine international journals. He served as the PC members of international conferences such as ACM Multimedia, ICME, and ICMR. He is the associate editor of Neurocomputing and KSII Transactions on Internet and Information Systems.

Yi Yang is a Senior lecturer of Computer Science with the Centre for Quantum Computation & Intelligent Systems, University of Technology, Sydney. Prior to that, he was a postdoc research fellow with the school of computer science, Carnegie Mellon University. He received the PhD degree in Computer Science from Zhejiang University in 2010. His research interest includes machine learning and its application to computer vision and multimedia analysis.