Retargeting Motion to New Characters By Dr. Michael Gleicher Aaron CrickenbergerCS551 Spring ‘02.

Slides:



Advertisements
Similar presentations
Configuration Space. Recap Represent environments as graphs –Paths are connected vertices –Make assumption that robot is a point Need to be able to use.
Advertisements

LOCOMOTION IN INTERACTIVE ENVIRONMENTS Navjot Garg.
Moused 2; The return of big blob.. MOUSE, MOUSE! We need your help. Big Blob didn’t die, and he is back. Before the explosion happened, he teleported away,
Layered Acting for Character Animation By Mira Dontcheva Gary Yngve Zoran Popović presented by Danny House SIGGRAPH 2003.
Video Game Math: Circle-on-Circle collision detection.
Verbs and Adverbs: Multidimensional Motion Interpolation Using Radial Basis Functions Presented by Sean Jellish Charles Rose Michael F. Cohen Bobby Bodenheimer.
Animation Following “Advanced Animation and Rendering Techniques” (chapter 15+16) By Agata Przybyszewska.
Forward and Inverse Kinematics CSE 3541 Matt Boggus.
A Momentum-based Bipedal Balance Controller Yuting Ye May 10, 2006.
Dr. S.M. Malaek Assistant: M. Younesi
Splines II – Interpolating Curves
Interactive Motion Editing Presented by Troy McMahon.
1Notes  Handing assignment 0 back (at the front of the room)  Read the newsgroup!  Planning to put 16mm films on the web soon (possibly tomorrow)
Physically Based Motion Transformation Zoran Popović Andrew Witkin SIGGRAPH ‘99.
Motion Editing and Retargetting Jinxiang Chai. Outline Motion editing [video, click here]here Motion retargeting [video, click here]here.
CSCE 441 Computer Graphics: Keyframe Animation/Smooth Curves Jinxiang Chai.
UNC Chapel Hill M. C. Lin Reading Assignments Principles of Traditional Animation Applied to 3D Computer Animation, by J. Lasseter, Proc. of ACM SIGGRAPH.
Animation by Adaptation Tutorial 2: Motion Editing Michael Gleicher Graphics Group Department of Computer Sciences University of Wisconsin – Madison
1 Expression Cloning Jung-yong Noh Ulrich Neumann Siggraph01.
CS274 Spring 01 Lecture 5 Copyright © Mark Meyer Lecture V Higher Level Motion Control CS274: Computer Animation and Simulation.
Retargetting Motion to New Characters Michael Gleicher SIGGRAPH 98.
Animation. Outline  Key frame animation  Hierarchical animation  Inverse kinematics.
Tracking with Linear Dynamic Models. Introduction Tracking is the problem of generating an inference about the motion of an object given a sequence of.
1cs426-winter-2008 Notes  SIGGRAPH crunch time - my apologies :-)
Binary “There are 10 types of people in the world… those that understand binary and those that don’t.”
Unit 1 – Improving Productivity Jake Carey. 1.1Why did you use a computer? What other systems / resources could you have used? I used a computer to finish.
Faking Dynamics of Cloth Animation for Animated Films Fabian Di Fiore Expertise Centre for Digital Media Hasselt University, Belgium
Chapter 3 Vectors Coordinate Systems Used to describe the position of a point in space Coordinate system consists of A fixed reference point called.
Introduction Tracking the corners Camera model and collision detection Keyframes Path Correction Controlling the entire path of a virtual camera In computer.
Inverse Kinematics Kris Hauser
CSE 380 – Computer Game Programming Box2D Box2D TestBed.
CSC418 Computer Graphics n Animation Principles n Keyframe Animation.
Spacetime Constraints Revisited Joe Marks J. Thomas Ngo Using genetic algorithms to find solutions to spacetime constraint problems in 2D.
Week 5 - Wednesday.  What did we talk about last time?  Exam 1!  And before that?  Review!  And before that?  if and switch statements.
hoe Last viewed 1 PowerPoint Slide Show (.pps) You can advance through each part of the screen by left clicking When you see the at the top right of the.
Motion Editing (Geometric and Constraint-Based Methods) Jehee Lee.
Computer Vision - A Modern Approach Set: Tracking Slides by D.A. Forsyth The three main issues in tracking.
Computer Graphics 2 In the name of God. Outline Introduction Animation The most important senior groups Animation techniques Summary Walking, running,…examples.
University of Texas at Austin CS384G - Computer Graphics Fall 2008 Don Fussell Parametric surfaces.
Flash! Macromedia Flash is the key to designing and delivering low-bandwidth animations, presentations, and Web sites. It offers scripting capabilities.
110/27/ :47 Graphics II Animation Introduction and Motion Control Session 6.
Retargetting Motion to New Characters Michael Gleicher.
CSCE 441 Computer Graphics: Keyframe Animation/Smooth Curves Jinxiang Chai.
1cs426-winter-2008 Notes  Will add references to splines on web page.
SEM Basics 2 Byrne Chapter 2 Kline pg 7-15, 50-51, ,
CSCE 441 Computer Graphics: Keyframe Animation/Smooth Curves Jinxiang Chai.
CS 450: COMPUTER GRAPHICS ANIMATION SPRING 2015 DR. MICHAEL J. REALE.
1 Project designed and created by M. Shajith Kumar.
Rick Parent - CIS681 Motion Capture Use digitized motion to animate a character.
CSCE 441: Keyframe Animation/Smooth Curves (Cont.) Jinxiang Chai.
CSCE 441: Keyframe Animation/Smooth Curves (Cont.) Jinxiang Chai.
INTRODUCTION TO DYNAMICS ANALYSIS OF ROBOTS (Part 4)
CSC321 Lecture 5 Applying backpropagation to shape recognition Geoffrey Hinton.
Motion Graphs By Lucas Kovar, Michael Gleicher, and Frederic Pighin Presented by Phil Harton.
Animation From Observation: Motion Editing Dan Kong CMPS 260 Final Project.
Transformations for GCSE Maths Enlargement Translation Reflection Rotation.
Animation Animation is about bringing things to life Technically: –Generate a sequence of images that, when played one after the other, make things move.
End effector End effector - the last coordinate system of figure Located in joint N. But usually, we want to specify it in base coordinates. 1.
LECTURE 17: BEYOND LINEARITY PT. 2 March 30, 2016 SDS 293 Machine Learning.
Computer Animation CS 446 September 5, 2001.
Character Animation Forward and Inverse Kinematics
CSCE 441 Computer Graphics: Keyframe Animation/Smooth Curves
WORKSHOP 9 BRAKE SYSTEM II
CSCE 441 Computer Graphics: Keyframe Animation/Smooth Curves
Reading Assignments Principles of Traditional Animation Applied to 3D Computer Animation, by J. Lasseter, Proc. of ACM SIGGRAPH 1987 Computer Animation:
Turning to the Masters: Motion Capturing Cartoons
OS – Memory Deallocation
Synthesis of Motion from Simple Animations
Computer Graphics Lecture 15.
Arrays.
Presentation transcript:

Retargeting Motion to New Characters By Dr. Michael Gleicher Aaron CrickenbergerCS551 Spring ‘02

Introduction Adapting a motion from one character to another –Characters are of the same form (in this case, all are bipeds) –Differ in terms of proportions and dimensions IK is used to some degree but really, it’s… Yet Another Spacetime Constraints Paper

Why do this? Motion capture “Humanize” creatures Slight changes have been made to character after it has already been animated Come on.. IK, frequency domain, spacetime constraints, B-splines… why wouldn’t Brogan love this?

What’s been done before? Character Studio adjusts keyframes to maintain footplants & balance Adjusting physical parameters to create a new motion for a new character (whereas we are now looking to re-use the old motion with a new character) Motion in the frequency domain

What’s been done before? Spacetime constraints allowed motions to be adjusted for the same character to do new things. A paper was written (Spacetime Constraints Revisited) and wasn’t really about spacetime constraints at all.

Mathy stuff q t – the configuration of our character (at time t) m(t) = m o (t) + d(t) –Retargeted motion is the original motion plus differences f i (t) – constraints on the motion –Footplants –Contact with a box

Getting it right How to deal with skating or intersecting joints? Let’s try IK! Whoops… IK will give us jittery motion. But some of the jittery motion may already have been there… OK, so let’s IK and then lowpass-filter our motion! Well crap, now we’re violating constraints again…

The big picture The problem: up until now, considering individual frames The solution: consider them all at once How? Spacetime constraints!

We’ll never get tired of them Generally define a constraint as: –f(q t i, q t j )  c,  ={>,<,=} Examples: –Parameter’s value in a range –Point on character in a specified location or region –Point stays in the same location or region for specified time –Point is a specified distance from another point –Vector between two points has a specified orientation

But math still sucks We can’t really define “grace” as a constraint The objective function (variance over the whole motion) doesn’t consider the motion as a whole (doesn’t want to waste motion to prepare to meet a goal). Oh heck, let’s use it anyway and hack around elsewhere to fix this…

B-Splines Restrict how close the control points for B- splines can be, and you indirectly control the highest possible frequency in your motion. Express d(t) as B-splines to smooth it out, but the original motion can still have snaps to it (ie: karate kick)

But wait! There’s more! We need to scale the motion just as we have scaled the figure… Now we may also need to translate the motion to put the character back where we want them We’ll use interpolation to get rid of high- frequencies caused by adding these changes to d(t)

Step by step 1.Start with initial motion, initial constraints 2.Translate/scale the motion w/ interpolation 3.Frequency decomposition 4.Non-linear spacetime constraints for a displacement that when added to step 2 gives us what we want. 5.If we’re not there, use the result of step 4 as the input to step 1.

What’s good? This looks at an entire motion, and so is able to create anticipated motions. –Eg: person slowly reaching for a box before they get to it It’s computed in near-real time, which allows the animator to tweak constraints to Get It Right

What’s bad? When it comes to constraints, limited just to “things we can solve.” Physics isn’t taken into account, so things could look unnatural

Pretty things

Yeah that’s pretty much it