Vous êtes ici

Human Motion Seminar : Motion Analysis and Shape Morphing

Vendredi, 18. novembre 2016 -
10:30 - 12:00
Zhiguand Liu, Hubert Shum
Salle Les Minquiers, Irisa
À l'occasion de la visite d'Hubert Shum de l'université de Northumbria (UK), Ludovic Hoyet et l'équipe MimeTIC organisent un séminaire vendredi 18 novembre 2016, 10h30-12h00 en salle Les Minquiers.
Human Motion Seminar : Motion Analysis and Shape Morphing
Dr. Zhiguand Liu (postdoc dans l'équipe MimeTIC arrivé début novembre)
Title: High-Quality Compatible Triangulation with Inconsistent Rotation and Shape Self-occlusion Enhancement for Planar Shape Morphing
Abstract: Planar shape morphing, also known as metamorphosis or shape blending, is the gradual transformation of one shape into another. Shape morphing techniques have been used widely in animation and special effects packages, such as Adobe After Effects and HTML5. With these morphing methods, we can transform a human into a bird or some other objects that people may never experience in real life. Thus, we want to build an interactive system that blends the human silhouette and other shapes such that the users can see these interesting transformations. To build such a system, (1) we need to employ compatible triangulation method to compute the correspondence between two shapes. (2) we need to apply shape interpolation method to transform one shape to another. (3) we need to use posture reconstruction method to address the transformation that involves self-occlusion.
Computing compatible triangulation can build the one-to-one correspondence between both the boundary and interior of two shapes. We propose
a new method to compute compatible triangulation of two polygons in order to create a smooth geometric transformation between them. We also present an efficient scheme to fix the inconsistent rotation problems that rigid shape interpolation algorithm may suffer. Lastly, we propose a new real-time probabilistic framework to enhance the accuracy of live captured postures that belong to one of the action classes in the database, which can be used to handle shapes with self-occlusion.
Bio: Zhiguang Liu received the PhD degree in computer science from the City University of Hong Kong in 2016. His research interests include character animation and machine learning.
Dr. Hubert Shum (Associate Professor à l'universtié de Northumbria, UK)
Title: Human Motion Understanding

Abstract: Due to the recent advancement in motion capture hardware and motion-based applications, human motion analysis has become an increasingly popular research area. Its core problem is to model human motion in a meaningful way, such that we can generalize knowledge to recognize, analyze and synthesize movement. Motion related applications nowadays such as motion-based gaming, 3D character animation, autonomous surveillance and smart robots are the results of the area.
The problem of human motion analysis is important as it connects different research fields. Taking an example of motion gaming with the Microsoft Kinect, the system first applies computer vision techniques to identify human body parts. Then, artificial intelligence is introduced to understand the meaning of the movement and perform human-computer interaction. Virtual reality techniques based on movement are sometimes used to enhance gaming immersiveness. Character animation and graphical rendering algorithms are implemented to render the controlled virtual character.
In this talk, I will discuss on the importance of human motion analysis in computer science. With the support of my research projects, I will demonstrate how motion analysis can connect different research fields, including computer graphics, games and vision. I will show how my projects achieve impact in research and the society, and conclude my presentation with future opportunities and potential directions.

Bio: Hubert P. H. Shum is an Associate Professor (Reader) in Computer Science and the Programme Leader of BSc (Hons) Computer Animation and Visual Effects at Northumbria University. He leads a research team focusing on computer graphics and computer vision, utilizing and managing the Motion Capture and Virtual Reality Laboratory. Before this, he worked as a Senior Lecturer at Northumbria University, a Lecturer at the University of Worcester, a postdoctoral researcher at RIKEN Japan, and a research assistant at the City University of Hong Kong. He received his PhD degree from the School of Informatics at the University of Edinburgh, as well as his Master and Bachelor degrees from the City University of Hong Kong. He has received £124,000 from EPSRC for a project on human motion analysis, and has been a core contributing researcher in a €3.03 million Erasmus Mundus project. On top of these, he has received more than £210,000 from Northumbria University to hire PhD students and purchase research equipment.