Monday, 8 August 9:00 am - 10:30 am | East Building, Ballroom C
Session Chair: Karen Liu, Georgia Institute of Technology
Physically Valid Statistical Models for Human Motion Generation
This paper shows how statistical modeling techniques can be combined with physics-based modeling techniques to address the limitations of both techniques. The power of this new model is generated by a wide range of physically realistic motions that achieve goals specified by the users.
Xiaolin Wei
Texas A&M University
Jianyuan Min
Texas A&M University
Jinxiang Chai
Texas A&M University
Motion Capture From Body-Mounted Cameras
A wearable motion capture system that uses mounted cameras in natural environments.
Takaaki Shiratori
Disney Research Pittsburgh
Hyun Soo Park
Carnegie Mellon University
Leonid Sigal
Disney Research Pittsburgh
Yaser Sheikh
Carnegie Mellon University
Jessica Hodgins
Carnegie Mellon University and Disney Research Pittsburgh
Motion Recons***ction Using Sparse Accelerometer Data
A novel framework for generating full-body animations controlled by only four 3D accelerometers that are attached to the extremities of a human actor. This approach relies on a knowledge base that consists of a large number of motion clips obtained from marker-based motion capturing.
Jochen Tautges
Rheinische Friedrich-Wilhelms-Universität Bonn
Arno Zinke
GfaR mbH
Björn Krüger
Rheinische Friedrich-Wilhelms-Universität Bonn
Jan Baumann
Rheinische Friedrich-Wilhelms-Universität Bonn
Andreas Weber
Rheinische Friedrich-Wilhelms-Universität Bonn
Thomas Helten
Universität des Saarlandes und Max-Planck-Institut für Informatik
Meinard Müller
Universität des Saarlandes und Max-Planck-Institut für Informatik
Hans-Peter Seidel
Universität des Saarlandes und Max-Planck-Institut für Informatik
Bernd Eberhardt
Hochschule der Medien Stuttgart
Video-Based Characters: Creating New Human Performances From a Multi-View Video Database
This paper presents a new data-driven approach for image-based synthesis of realistic video animations containing user-defined human motions seen from user-defined camera views.
Feng Xu