|
Biomechanical Modeling and Control of the Human Body
Sung-Hee Lee, Eftychios Sifakis, Jinho Park, Demetri Terzopoulos
We develop a comprehensive biomechanical model of the human body, confronting the combined challenge of modeling and controlling more or less all of the relevant articular bones and muscles, as well as simulating the physics-based deformation of the soft tissues.
|
|
Embedded Sketch Recognition
Gabriele Nataneli
We devised a sketch recognition pipeline that is very flexible and
comes with minimal computation requirements. We implemented this framework on a
variety of devices including the Nintendo DS and the iPhone.
|
|
Egocentric Affordance Fields in Pedestrian Steering
Mubbasir Kapadia, Shawn Singh, Billy Hewlett
We present a system for local path-planning and steering that can be easily extended to perform high level group behaviors. Our framework is based on the concept of affordances the possible ways an agent can interact with its environment. Each agent perceives the environment through a set of fields that are represented in the agent s local space, termed as Egocentric Perception Fields. Using our framework, we demonstrate autonomous virtual pedestrians that perform steering and path planning in unknown environments along with the emergence of high level
responses to never seen before situations.
|
|
AI Fighter
Billy Hewlett
In this fast paced fighting game, you and your human opponent compete with
by fighting each other's students. Each student AI, in turn, learns how to fight
by observing its human teacher fight her opponent's student. The winner is the
player who can teach their student to beat the other player first. Read about and
download this PC game here
|
|
Partially Precomputed A*
Billy Hewlett
It is computationally expensive to perform A* on a large number of agents in
real time applications. Our work is an amoritized search algorithm which can
outperform A* for large numbers of agents searching the same graph.
|
|
Controller Synthesis for Physically Based Character Animation
Brian Allen
The goal of this project is to develop a machine learning system to
automatically synthesize real-time actuator controllers for physically
simulated characters. Such a system should ideally rely on only the mechanical
structure of the character and a high-level description of the desired
behavior. Bipedal locomotion for human-scale characters has been used to
validate the approach.
|
|
Automatic Splicing for Hand and Body Animations
Anna Majkowska
We propose a solution to a new problem in animation research: how to use human
motion capture data to create character motion with detailed hand gesticulation
without the need for the simultaneous capture of hands and the full body.
|
|
Sketch Based Facial Animation
Gabriele Nataneli
We are developing a sketch-based interface for driving facial animation.
Hand-drawn sketches are the most natural medium for artists and children to
express their visions; our goal is to conceal the complexity of a facial
animation system through an interface which is both powerful and enjoyable to
use.
|
|
Character Skinning
Albert Chu
We are developing a character skinning interface module for the D.A.N.C.E
software system.
|
|
The Virtual Stuntman
Ari Shapiro
We are developing a multi-tasking, autonomous virtual stuntman using a
combination of physics-based and kinematic techniques.
|
|
Speech To Facial Animation
Yong Cao
We are developing a system for speech-to-facial animation in collaboration with
Dr. Fred Pighin at ICT/USC.
|
|
Constrained Simulation of Particle Systems and Rigid Bodies
Lucio Flores
We are implementing and modifying existing techniques for the physics-based
simulation of particle and rigid bodies. Eventually the project will develop a
simulation engine for articulated bodies. We are investigating methods based on
cartesian and reduced coordinate systems.
|
|
Real Time Cloth Simulation
Ray Huang, Alan Gasperini
We are working on real-time cloth simulation based on a technique presented on
the paper "Advanced Character Physics" by Jakobsen and Thomas.
|