SLIDE 1
CS4496 Computer Animation
Instructor: C. Karen Liu
Karen Liu
Associate Professor at School of Interactive Computing
CS4496 Computer Animation Instructor: C. Karen Liu Karen Liu - - PowerPoint PPT Presentation
CS4496 Computer Animation Instructor: C. Karen Liu Karen Liu - - PowerPoint PPT Presentation
CS4496 Computer Animation Instructor: C. Karen Liu Karen Liu Associate Professor at School of Interactive Computing Karen Liu Associate Professor at School of Interactive Computing Our TA: John Turner Administrations
SLIDE 2
SLIDE 3
Karen Liu
Associate Professor at School of Interactive Computing
SLIDE 4
Our TA: John Turner
SLIDE 5
Administrations
✦ http://www.cc.gatech.edu/classes/AY2015/cs4496_spring/ ✦ Course slides and projects are online. ✦ No textbook, but there will be online reading materials. ✦ Office hours: Mon 12-00-1:00, TSRB 230A. ✦ TA office hours: Wed 12:30-2:30, COC at the tables.
SLIDE 6
My goals
✦ Give you an overview of computer animation with an emphasis on
physics-based animation and character animation.
✦ Teach you how to be a good engineer who also understands art. ✦ Inspire some of you to do research in computer animation.
SLIDE 7
You will learn how to…
✦ Interpolate keyframes using different families of splines. ✦ Simulate particles, rigid bodies, articulated rigid bodies,
deformable objects, fluids, cloth.
✦ Numerically integrate an ODEs ✦ Implement constrained systems, collision detection and handling,
inverse kinematics, and grid-based fluid solvers.
✦ Synthesize animal locomotion, hand manipulation, facial
animation, skin/muscle deformation.
SLIDE 8
You will not learn…
✦ How to use commercial animation tools such as Maya (maybe a
little bit).
✦ How to use game engines such as Unity. ✦ How to render high quality animation.
SLIDE 9
Prerequisites
✦ Thorough understanding of linear algebra. ✦ Vector calculus. ✦ A good working knowledge of C and C++ programming.
SLIDE 10
Grading
✦ Reading assignments (0%) ✦ Six projects (75%) ✦ Final exam (25%)
SLIDE 11
Projects
✦ Project 1: Splines. ✦ Project 2: Gallio’s experiment. ✦ Project 3: Tinkertoy ✦ Project 4: Rigid body smoothie. ✦ Project 5: Fun with fluids. ✦ Project 6: Twister. ✦ Late policy: 20% reduction per day if you don’t have a good reason. ✦ Everything has to be turned in before 10 am on the due date.
SLIDE 12
Quiz
✦ Quiz will be given and discussed in class from time to time. ✦ For every topic, expect 2-4 problems.
SLIDE 13
Quiz
✦ Given two vectors, a = (3, 0, 1) and b = (-2, 5, 2), ✦ What is the dot product of a and b? ✦ What is the cross product of a and b? ✦ What is the norm of a? ✦ What is the angle between a and b? ✦ What is the projection of a on b?
SLIDE 14
Quiz
✦ Given three n by n matrices A, B, and C ✦ Is AB = BA true? ✦ Is A(BC) = (AB)C true? ✦ Is (AB)T = ATBT true? ✦ Does A-1 always exist? ✦ What is the rank of A?
SLIDE 15
About Maya
✦ You need to bring a laptop with Maya installed next Wednesday. ✦ No prior knowledge in Maya is required. ✦ Simple Maya tutorial to help you start. ✦ http://students.autodesk.com.
SLIDE 16
About DART
✦ A simulation framework you will use to build your projects on. ✦ Skeleton code is provided for each project so you don’t need to start
from scratch.
✦ Everything is written in C++ ✦ Source code is available on Github.
SLIDE 17
Introduction
✦ Keyframe animation ✦ Physics simulation. ✦ Character animation.
SLIDE 18
Traditional animation
✦ Film runs at 24 frames per sec; that is, 1440
pictures to draw in one minute.
✦ Artistic vision has to be converted into a
sequence of still “keyframes”.
✦ Hard to draw consistent “in-between”
frames.
✦ Not enough to get the still right; must to
look right at full speed.
SLIDE 19
Computer assisted animation
✦ Generate the images by rendering a
3D model.
✦ Manually set the parameters for
each keyframe.
✦ Automatically interpolate between
two drawings to produce inbetweens.
SLIDE 20
Did it really get better?
Pencil and paper Computer Do computers really expedite the process of creating animation?
SLIDE 21
What can’t be done by keyframes?
SLIDE 22
Introduction
✦ Keyframe animation ✦ Physics simulation. ✦ Character animation.
SLIDE 23
Physics simulation
✦ An algorithm that produces a sequence of states over time under
the laws of physics.
✦ What is a state?
SLIDE 24
Simulation
xi ∆x xi+1 xi+1 = xi + ∆x xi
SLIDE 25
Simulation
xi ∆x xi+1 xi+1 = xi + ∆x xi
Newtonian laws gravity wind gust elastic force… integrator
SLIDE 26
Ordinary differential equations
˙ x(t) = f(x(t))
known function time derivative of the unknown function unknown function that evaluates the state given time
F = ma An ODE is an equation involving a function and its derivatives.
SLIDE 27
Quiz
✦ What function does the black box represent?
x ˙ x f
1. 2. 3.
SLIDE 28
Solving ODE
✦ Standard introductory differential equation courses focus on
finding solutions analytically.
✦ Linear ODEs can be solved by integral transforms. ✦ We will learn how to solve an ODE numerically in this class.
SLIDE 29
Rigid bodies
SLIDE 30
Rigid bodies
SLIDE 31
Articulated rigid bodies
SLIDE 32
Deformable bodies
SLIDE 33
Deformable bodies
SLIDE 34
Plastic materials
SLIDE 35
Fluids
SLIDE 36
Fluids
SLIDE 37
Fluids
SLIDE 38
Fluids
SLIDE 39
Cloth
SLIDE 40
Cloth
SLIDE 41
Fluids + Solids
SLIDE 42
Fluids + Cloth
SLIDE 43
Melted
SLIDE 44
Drowned
SLIDE 45
Immolated
SLIDE 46
Waterboarded
SLIDE 47
Skewered
SLIDE 48
Crushed
SLIDE 49
Mutilated
SLIDE 50
Introduction
✦ Keyframe animation ✦ Physics simulation. ✦ Character animation.
SLIDE 51
Control rigid bodies
Popovic et al, 2001
SLIDE 52
Control rigid bodies
SLIDE 53
Control rigid bodies
SLIDE 54
Control fluids
SLIDE 55
Control fluids
SLIDE 56
Control virtual humans
SLIDE 57
Why is this hard?
✦ Human musculoskeletal system has a large number of degrees of
freedom and nonlinear dynamics
✦ The control mechanism must be robust to operate in complex
world.
✦ No qualitative ways to measure “Naturalness”
SLIDE 58
Human motion
SLIDE 59
Without control
SLIDE 60
Why is this hard?
✦ Human musculoskeletal system has a large number of degrees of
freedom and nonlinear dynamics
✦ The control mechanism must be robust to operate in complex
world.
✦ No qualitative ways to measure “Naturalness”
SLIDE 61
Complex animals movements
SLIDE 62
Simulation
xi ∆x xi+1 xi+1 = xi + ∆x xi
Newtonian laws gravity wind gust elastic force… integrator
SLIDE 63
Simulation
xi ∆x xi+1
Newtonian laws gravity wind gust elastic force… integrator
SLIDE 64
Simulation + Control
xi ∆x xi+1
Newtonian laws integrator gravity contact force muscle force…
SLIDE 65
Physics approach
SLIDE 66
Physics approach
SLIDE 67
Physics approach
SLIDE 68
Why is this hard?
✦ Human musculoskeletal system has a large number of degrees of
freedom and nonlinear dynamics
✦ The control mechanism must be robust to operate in complex
world.
✦ No qualitative ways to measure “Naturalness”
SLIDE 69
Data-driven approach
✦ Use captured data to synthesize
and interpolate new motions.
✦ Preserve detailed nuances of
human movements in the real world.
✦ Produce stable and compelling
animations in real-time.
SLIDE 70
Motion capture
✦ Many different approaches. Our lab has an
- ptical based system.
✦ Multiple infrared cameras. ✦ High temporal resolution (1000+ fps). ✦ Detect the locations of reflective markers.
SLIDE 71
Raw data from mocap
✦ Capture 3D positions of the markers. ✦ Require inverse kinematics post-processing to produce animation
represented as joint angle trajectories.
SLIDE 72
Final motion
SLIDE 73
Skinning
SLIDE 74