Syllabus
This course will discuss the state-of-the-art methods for synthesis of
realistic computer animation. It covers the basic computational and
physical methods for modeling naturally occurring motion, as well as the
advanced techniques for character and secondary motion synthesis. In
additional to synthetic approaches to computer animation, the course will also
explore the methods which transform motion acquired from real actors. The
final course project is aimed at extending the current research efforts in novel
directions.
Administrivia
- Time: TTh 12-1:20
- Place: EE1 003
- Instructor:
Zoran Popovic (zoran@cs), Sieg 325B
- Office hours: Tue 11:00am-12:00pm
Prerequisites
- CSE557 or equivalent
- Thorough understanding of linear algebra
- Vector calculus
- A good working knowledge of C (and hopefully C++) programming
Text
- Text supplements will be provided in class.
Topics
- Differential equations
- Rigid-body simulation
- Collision detection and response
- Inverse control of simulations
- Automatic differentiation
- Inverse kinematics
- Numerical methods for continuous optimization
- Spacetime optimization
- Motion editing
- Robot controllers
- Data-based models for character animation
- Machine learning and animation
- Fluid dynamics
- Cloth
- Fracture, explosions
- System identification
- Physics and vision
Grading
- Project 1: 25%
- Project 2: 30%
- Project 3: 45%
There will be no final exam for this course.
Projects
Students have the option of using the graphics
instructional lab for their project work. The lab consists of 14 Intel machines running
NT, and it is located in Sieg 228.
Here is a tentative list of projects:
- Project 1: Constrained particles
- An exercise in differential equations, and constrained particle systems.
-
- Project 2: Inverse kinematics
- Differential solution to the problem of robustly computing joint angles
from the intermittent motion capture data. The fastest and most robust
implementation wins a prize.
-
- Project 3: Final project
- This project is intended to be research exploration.
Projects should be done in groups of two. Depending on the size of the
problem, the final project groups may include 3 people. Projects will be graded during
the in-person sessions with the instructor. Each member of the team should be
able to answer questions on all aspects of the project.
Project Turn-in & Late Policy
Projects should be demo-ed during the allotted time on the due date.
Exceptions will be given only in extreme circumstances and only
in advance.