Courses

Wednesday, 15 December | 11:00 AM - 12:45 PM | Room 308A

Presented in English / 영어로 발표 됨

Introduction to Data-Driven Animation: Programming With Motion Capture

Tuesday, 14 December | 9:00 午後 - 10:45 午後 | Room 308A

Data-driven animation using motion-capture data has become a standard practice in character animation. A number of techniques have been developed to add flexibility on captured human motion data by editing joint trajectories, warping motion paths, blending a family of parameterized motions, splicing motion segments, and adapting motion to new characters and environments.

Even with the abundance of motion-capture data and the popularity of data-driven animation techniques, programming with motion capture-data is still not easy. A single clip of motion data encompasses a lot of heterogeneous information including joint angles, the position and orientation of the skeletal root, their temporal trajectories, and a number of coordinate systems. Due to this complexity, even simple operations on motion data, such as linear interpolation, are rarely described as succinct mathematical equations in research papers.

This course provides not only a solid mathematical background, but also a practical guide to programming with motion-capture data. It begins with a brief review of affine geometry and coordinate-invariant (conventionally called coordinate-free) geometric programming, which generalizes incrementally to deal with three-dimensional rotations/orientations, the poses of an articulated figure, and full-body motion data. Then it identifies a collection of coordinate-invariant operations on full-body motion data and their object-oriented implementation. Finally, it explains the practical use of this programming framework in a variety of contexts ranging from data-driven manipulation and interpolation to state-of-the-art biped locomotion control.

Level

Beginner

Intended Audience

Mainly beginners who are interested in data-driven animation and motion capture, and experienced programmers who need to animate articulated characters. Advanced researchers in computer animation may also benefit from the mathematical framework and programming style presented in the course.

Presentation Language

Presented in English / 영어로 발표 됨

Prerequisites

Knowledge of undergraduate-level computer graphics.

Syllabus

Introduction and Overview
  Data-driven animation using motion-capture data
  Why is it difficult to do programming with motion-capture data?
  Course overview

Coordinate-Invariant Programming with Points and Vectors
  What is coordinate-invariant geometric programming?
  Affine geometry
  Coordinate-invariant operations between points and vectors

Programming with Orientations and Rotations
  Representing orientations and rotations
  Analogy between points/vectors and orientations/rotations
  Coordinate-invariant operations with orientations and rotations
  Linear combination of rotations
  Affine combination of orientations

Programming with Motion Capture Data
  Representing the poses and pose displacements of an articulated figure
  Motion data and displacements
  Coordinate-invariant programming with motion data

Practical Examples
  Distance measure
  Hierarchical displacement mapping
  Interpolation and transitioning
  Style transfer
  Feedback control for physically based biped locomotion

Jehee Lee
Seoul National University

Instructor Bios

Jehee Lee
Jehee Lee is an associate professor at Seoul National University. He received his BS, MS, and PhD in computer science from Korea Advanced Institute of Science and Technology in 1993, 1995, and 2000, respectively. He leads the SNU Movement Research Laboratory, where his research interests in computer graphics and animation are focused on developing new ways of understanding, representing, and animating human movement. This involves full-body motion analysis and synthesis, biped control and simulation, motion capture, motion planning, data-driven and physically based techniques, interactive avatar control, crowd simulation, and facial animation.