1. CS559: Computer Graphics Lecture 36: Animation Li Zhang Spring 2008 Slides from Brian Curless at U of Washington

2. Today Particle Systems, Cartoon animation, ray tracing Reading – (Optional) John Lasseter. Principles of traditional animation applied to 3D computer animation. Proceedings of SIGGRAPH (Computer Graphics) 21(4): 35-44, July 1987. http://portal.acm.org/citation.cfm?id=37407 – (Optional) WILLIAM T. REEVES, ACM Transactions on Graphics, Vol. 2, No. 2, April 1983 http://portal.acm.org/citation.cfm?id=357320

3. Particle system diff. eq. solver We can solve the evolution of a particle system again using the Euler method: void EulerStep(ParticleSystem p, float DeltaT){ ParticleDeriv(p,temp1); /* get deriv */ ScaleVector(temp1,DeltaT) /* scale it */ ParticleGetState(p,temp2); /* get state */ AddVectors(temp1,temp2,temp2); /* add -> temp2 */ ParticleSetState(p,temp2); /* update state */ p->t += DeltaT; /* update time */ } void EulerStep(ParticleSystem p, float DeltaT){ ParticleDeriv(p,temp1); /* get deriv */ ScaleVector(temp1,DeltaT) /* scale it */ ParticleGetState(p,temp2); /* get state */ AddVectors(temp1,temp2,temp2); /* add -> temp2 */ ParticleSetState(p,temp2); /* update state */ p->t += DeltaT; /* update time */ }

4. Bouncing off the walls Handling collisions is a useful add-on for a particle simulator. For now, we’ll just consider simple point-plane collisions. A plane is fully specified by any point P on the plane and its normal N. N P v x

5. Collision Detection How do you decide when you’ve made exact contact with the plane? N P v x

6. Normal and tangential velocity To compute the collision response, we need to consider the normal and tangential components of a particle’s velocity. N P v x v

7. Collision Response before after v’v’ The response to collision is then to immediately replace the current velocity with a new velocity: The particle will then move according to this velocity in the next timestep. v

8. Collision without contact In general, we don’t sample moments in time when particles are in exact contact with the surface. There are a variety of ways to deal with this problem. A simple alternative is to determine if a collision must have occurred in the past, and then pretend that you’re currently in exact contact.

9. Very simple collision response How do you decide when you’ve had a collision? A problem with this approach is that particles will disappear under the surface. Also, the response may not be enough to bring a particle to the other side of a wall. N P v1v1 x1x1 x2x2 x3x3 v2v2 v3v3

10. More complicated collision response Another solution is to modify the update scheme to: – detect the future time and point of collision – reflect the particle within the time-step N P v x

11. Generate Particles Particle Attributes – initial position, – initial velocity (both speed and direction), – initial size, – initial color, – initial transparency, – shape, – lifetime. WILLIAM T. REEVES, ACM Transactions on Graphics, Vol. 2, No. 2, April 1983

12. Generate Particles Particle Attributes – initial position, – initial velocity (both speed and direction), – initial size, – initial color, – initial transparency, – shape, – lifetime. WILLIAM T. REEVES, ACM Transactions on Graphics, Vol. 2, No. 2, April 1983

13. Generate Particles Particle Attributes – initial position, – initial velocity (both speed and direction), – initial size, – initial color, – initial transparency, – shape, – lifetime. WILLIAM T. REEVES, ACM Transactions on Graphics, Vol. 2, No. 2, April 1983

14. Generate Particles Initial Particle Distribution Particle hierarchy, for example – Skyrocket : firework – Clouds : water drops

15. Throwing a ball from a robot arm Let’s say we had our robot arm example and we wanted to launch particles from its tip. How would we calculate initial speed? Q=R(theta)*T1*R(phi)*T2*R(psi)*P We want dQ/dt

16. Principles of Animation John Lasseter. Principles of traditional animation applied to 3D computer animation. Proceedings of SIGGRAPH (Computer Graphics) 21(4): 35-44, July 1987. Goal : make characters that move in a convincing way to communicate personality and mood. Walt Disney developed a number of principles. – ~1930 Computer graphics animators have adapted them to 3D animation.

17. Principles of Animation The following are a set of principles to keep in mind: 1. Squash and stretch 2. Staging 3. Timing 4. Anticipation 5. Follow through 6. Secondary action 7. Straight-ahead vs. pose-to-pose vs. blocking 8. Arcs 9. Slow in, slow out 10. Exaggeration 11. Appeal

18. Squash and stretch Squash: flatten an object or character by pressure or by its own power. Stretch: used to increase the sense of speed and emphasize the squash by contrast. Note: keep volume constant! http://www.siggraph.org/education/materials/HyperGraph/animation/character_ animation/principles/squash_and_stretch.htm http://www.siggraph.org/education/materials/HyperGraph/animation/character_ animation/principles/squash_and_stretch.htm http://www.siggraph.org/education/materials/HyperGraph/animation/character_ animation/principles/bouncing_ball_example_of_slow_in_out.htm http://www.siggraph.org/education/materials/HyperGraph/animation/character_ animation/principles/bouncing_ball_example_of_slow_in_out.htm

19. Squash and stretch (cont’d)

21. Anticipation An action has three parts: anticipation, action, reaction. Anatomical motivation: a muscle must extend before it can contract. Watch: bugs-bunny.virtualdub.new.mpg Prepares audience for action so they know what to expect. Directs audience's attention.

22. Anticipation (cont’d) Amount of anticipation (combined with timing) can affect perception of speed or weight.

23. Arcs Avoid straight lines since most things in nature move in arcs.

24. Slow in and slow out An extreme pose can be emphasized by slowing down as you get to it (and as you leave it). In practice, many things do not move abruptly but start and stop gradually.

25. Exaggeration Get to the heart of the idea and emphasize it so the audience can see it.

26. Exaggeration Get to the heart of the idea and emphasize it so the audience can see it.

27. Appeal The character must interest the viewer. It doesn't have to be cute and cuddly. Design, simplicity, behavior all affect appeal. Example: Luxo, Jr. is made to appear childlike. http://www.youtube.com/watch?v=HDuRXvtImQ0&feature=related

28. Appeal (cont’d) Note: avoid perfect symmetries.

29. Appeal (cont’d) Note: avoid perfect symmetries.

30. Ray Tracing Slides are from Ravi Ramamoorthi’s graphics class at Columbia U Reading: Shirley Ch 10.1 --- 10.8

31. Effects needed for Realism  Reflections (Mirrors and Glossy)  Transparency (Water, Glass)  Interreflections (Color Bleeding)  (Soft) Shadows  Complex Illumination (Natural, Area Light)  Realistic Materials (Velvet, Paints, Glass)  And many more Image courtesy Paul Heckbert 1983

32. Ray Tracing  Different Approach to Image Synthesis as compared to Hardware pipeline (OpenGL)  OpenGL : Object by Object  Ray Tracing : Pixel by Pixel  Advantage:  Easy to compute shadows/transparency/etc  Disadvantage:  Slow (in early days)

33. Basic Version: Ray Casting Virtual Viewpoint Virtual ScreenObjects Ray misses all objects: Pixel colored blackRay intersects object: shade using color, lights, materialsMultiple intersections: Use closest one (as does OpenGL)

34. Ray Casting Produce same images as with OpenGL  Visibility per pixel instead of Z-buffer  Find nearest object by shooting rays into scene  Shade it as in standard OpenGL Section 10.1-10.2 in text (we show visually, omitting math)

35. Comparison to hardware scan-line  Per-pixel evaluation, per-pixel rays (not scan-convert each object). On face of it, costly  But good for walkthroughs of extremely large models (amortize preprocessing, low complexity)  More complex shading, lighting effects possible

36. Shadows Virtual Viewpoint Virtual ScreenObjects 10.5 in textbook Light Source Shadow ray to light is unblocked: object visibleShadow ray to light is blocked: object in shadow

37. Shadows: Numerical Issues Numerical inaccuracy may cause intersection to be below surface (effect exaggerated in figure) Causing surface to incorrectly shadow itself Move a little towards light before shooting shadow ray

38. Mirror Reflections/Refractions Virtual Viewpoint Virtual ScreenObjects 10.6 in textbook Generate reflected ray in mirror direction, Get reflections and refractions of objects

39. Recursive Ray Tracing (Core Idea) For each pixel  Trace Primary Eye Ray, find intersection  Trace Secondary Shadow Ray(s) to all light(s)  Color = Visible ? Illumination Model : 0 ;  Trace Reflected Ray  Color += reflectivity * Color of reflected ray  Trace Refracted Ray  Color += transparency * Color of refracted ray Also see section 10.4 in text Recursive function Calls

40. Problems with Recursion  Reflection rays may be traced forever  Generally, set maximum recursion depth

41. Turner Whitted 1980

42. Effects needed for Realism (Soft) Shadows Reflections (Mirrors and Glossy) Transparency (Water, Glass) Interreflections (Color Bleeding) Complex Illumination (Natural, Area Light) Realistic Materials (Velvet, Paints, Glass) Discussed in this lecture Not discussed so far but possible with distribution ray tracing (10.11) Hard (but not impossible) with ray tracing; radiosity methods

43. How to implement?  Ray parameterization  Ray-Surface Intersection

44. Ray/Object Intersections  Heart of Ray Tracer  One of the main initial research areas  Optimized routines for wide variety of primitives  Various types of info  Shadow rays: Intersection/No Intersection  Primary rays: Point of intersection, material, normals  Texture coordinates Section 10.3

45. Example  Sphere  How to decide there is an intersection?  Triangle  How to decide the intersection is inside?  Polygon  How to decide the intersection is inside?  How about an ellipsoid?

46. Ray-Tracing Transformed Objects We have an optimized ray-sphere test  But we want to ray trace an ellipsoid… Solution: Ellipsoid transforms sphere  Apply inverse transform to ray, use ray-sphere Section 10.8 of text

47. Acceleration Testing each object for each ray is slow  Faster Intersections  Optimized Ray-Object Intersections  Fewer Intersections Section 10.9 goes into more detail

48. Acceleration Structures Bounding boxes (possibly hierarchical) If no intersection bounding box, needn’t check objects Bounding Box Ray Spatial Hierarchies (Oct-trees, kd trees, BSP trees)

49. Octtree

50. K-d tree

51. Acceleration Structures: Grids

53. Raytracing on Graphics Hardware  Modern Programmable Hardware general streaming architecture  Can map various elements of ray tracing  Kernels like eye rays, intersect etc.  In vertex or fragment programs  Convergence between hardware, ray tracing [Purcell et al. 2002, 2003] http://graphics.stanford.edu/papers/photongfx

55. Outline  History  Basic Ray Casting (instead of rasterization)  Comparison to hardware scan conversion  Shadows / Reflections (core algorithm)  Ray-Surface Intersection  Optimizations  Current Research Section 10 in text

56. Ray Tracing: History  Appel 68  Whitted 80 [recursive ray tracing]  Landmark in computer graphics  Lots of work on various geometric primitives  Lots of work on accelerations  Current Research  Real-Time raytracing (historically, slow technique)  Ray tracing architecture

57. Outline  History  Basic Ray Casting (instead of rasterization)  Comparison to hardware scan conversion  Shadows / Reflections (core algorithm)  Ray-Surface Intersection  Optimizations  Current Research

58. Outline  History  Basic Ray Casting (instead of rasterization)  Comparison to hardware scan conversion  Shadows / Reflections (core algorithm)  Ray-Surface Intersection  Optimizations  Current Research

59. Outline  History  Basic Ray Casting (instead of rasterization)  Comparison to hardware scan conversion  Shadows / Reflections (core algorithm)  Ray-Surface Intersection  Optimizations  Current Research

60. Outline  History  Basic Ray Casting (instead of rasterization)  Comparison to hardware scan conversion  Shadows / Reflections (core algorithm)  Ray-Surface Intersection  Optimizations  Current Research

61. Interactive Raytracing  Ray tracing historically slow  Now viable alternative for complex scenes  Key is sublinear complexity with acceleration; need not process all triangles in scene  Allows many effects hard in hardware  OpenRT project real-time ray tracing (http://www.openrt.de)