Tuesday, 9 August 3:45 pm - 5:35 pm | East Building, Exhibit Hall A
Session Chair: KangKang Yin, National University of Singapore
Space-Time Planning With Parameterized Locomotion Controllers
A technique for animating characters traversing complex dynamic environments. The method uses motion capture to construct a set of controllers to traverse low-dimensional obstacles, and then employs a space-time planner to plan out the sequence in which the controllers must be used to traverse a complex dynamic environment.
Sergey Levine
Stanford University
Yongjoon Lee
University of Washington
Vladlen Koltun
Stanford University
Zoran Popović
University of Washington
Articulated Swimming Creatures
A general approach to creating realistic swimming behavior for a given articulated creature body, which includes creature/fluid simulation and swimming gait optimization. For a wide variety of aquatic creatures with different body shapes, the resulting motion matches that of real-world animals.
Jie Tan
Georgia Institute of Technology
Yuting Gu
Georgia Institute of Technology
Greg Turk
Georgia Institute of Technology
Karen Liu
Georgia Institute of Technology
Locomotion Skills for Simulated Quadrupeds
This paper presents a large number of integrated locomotion skills for real-time physics-based quadrupeds. These skills include walk, trot, pace, canter, gallop, sit, lie-down, get-up, and parameterized leaps over barriers, across gaps, and onto platforms. The simulated motions are compared to known data for a dog.
Stelian Coros
University of British Columbia
Andrej Karpathy
University of British Columbia
Ben Jones
University of British Columbia
Lionel Reveret
INRIA
Michiel van de Panne
University of British Columbia
Composite Control of Physically Simulated Characters
This approach to constructing composite controllers that track multiple trajectories in parallel generates both high-quality animations and natural responses to certain disturbances.
Uldarico Muico
University of Washington
Jovan Popović
Abobe Systems Incorporated
Zoran Popović
University of Washington
Character Animation in Two-Player Adversarial Games
This paper describes how game theory can be used to automatically create intelligent real-time controllers for characters engaged in a competitive game. These controllers employ long-term planning, display an intelligent use of randomness, and exhibit nuanced strategies based on unpredictability, such as feints and misdirection moves.
Kevin Wample