cama lesson 12 - inria

24
Computer Animation Lesson 12 - Ragdoll physics Remi Ronfard, Nov 2019

Upload: others

Post on 06-Jun-2022

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: CAMA LESSON 12 - Inria

Computer AnimationLesson 12 - Ragdoll physics

Remi Ronfard, Nov 2019

Page 2: CAMA LESSON 12 - Inria

12 Appeal

Page 3: CAMA LESSON 12 - Inria
Page 4: CAMA LESSON 12 - Inria

Ragdoll physics =« lifelike death »

Ragdoll physics is a type of procedural animation that is often used as a replacement for traditional static death animation. The term ragdoll comes from the problem that the articulated systems, due to the limits of the solvers used, tend to have little or zero joint stiffness, leading to a character collapsing much like a toy into comically improbable or compromising positions.

Page 5: CAMA LESSON 12 - Inria

Rag-doll physics

Page 6: CAMA LESSON 12 - Inria

Ragdoll physicsDead bodies -> all physics, no muscles

Ragdolls have been implemented using Featherstone's algorithm and spring-damper contacts (John Nagle, US Patent, 2000, Method and system for generating realistic collisions in graphical simulation). An alternative approach uses constraint solvers and idealized contacts (Physically Based Modeling Principles and Practice. Proc. SIGGRAPH '97. )

Page 7: CAMA LESSON 12 - Inria

Procedural animation of ragdollsPhysics for multi-layered physical models in non-playing characters (bones / muscle / nervous systems) and deformable scenic elements from "simulated materials" in vehicles, etc. Alternative to pre-made animation: each game is unique, whilst still deterministic. Rather than detract from gameplay through overstimulation, the "natural" qualities of movement provide for a more immersive experience, and extended replayability.

Page 8: CAMA LESSON 12 - Inria

Blended rag-dolls!Blended ragdoll works by playing a pre-made animation, but constraining the output of that animation to what a physical system would allow.

!Used in many games: ! Halo 2, Call of Duty 4, Left 4 Dead, Medal of

Honor Airborne, Unchartered Drake’s Fortune, etc.

Page 9: CAMA LESSON 12 - Inria

Blended Ragdoll Example : Running

Page 10: CAMA LESSON 12 - Inria

Blended Ragdoll Example : Running

Source: Kwon, Hodgins: Control Systems for Human Running using an Inverted Pendulum Model and a Reference Motion Capture Sequence. SCA 2010.

Page 11: CAMA LESSON 12 - Inria

Blended Ragdoll (continued)

Captured motion is divided into small segments

Page 12: CAMA LESSON 12 - Inria

Blended Ragdoll (continued)

Compute inverted pendulum model per segment

Page 13: CAMA LESSON 12 - Inria

Blended Ragdoll (continued)

Adapt segment to the chosen trajectory and context

Page 14: CAMA LESSON 12 - Inria

Quasi-physics

• Source: Sergey Levine and Jovan Popović. SCA 2012. Physically plausible simulation for character animation. 

Page 15: CAMA LESSON 12 - Inria

Quasi-physics

• Source: Sergey Levine and Jovan Popović. SCA 2012. Physically plausible simulation for character animation. 

Page 16: CAMA LESSON 12 - Inria

Quasi-physics

• Torque term

• Pose term

• End effector term

• Source: Sergey Levine and Jovan Popović. SCA 2012. Physically plausible simulation for character animation. 

Page 17: CAMA LESSON 12 - Inria

Quasi-physics

• Goal constraint

• Equations of motion

• Source: Sergey Levine and Jovan Popović. SCA 2012. Physically plausible simulation for character animation. 

Page 18: CAMA LESSON 12 - Inria

Quasi physics

Page 19: CAMA LESSON 12 - Inria

Biologically Inspired Control of Physically Simulated Bipeds

Source: Torsten Reil and Colm Massey, Theory Biosci. 2001.

Page 20: CAMA LESSON 12 - Inria

Biologically Inspired Control of Physically Simulated Bipeds

Page 21: CAMA LESSON 12 - Inria

Puppet Master

Page 22: CAMA LESSON 12 - Inria

Puppet Master

Page 23: CAMA LESSON 12 - Inria

Puppet Master

Page 24: CAMA LESSON 12 - Inria

Paper 12 - Deep Mimic