Lec12 ImageGeneration s06

Uploaded from authorPOINTLite
Views:
 
Category: Entertainment
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Light and Rendering: 

Light and Rendering CS5600 Computer Graphics Rich Riesenfeld Spring 2006 Lecture Set 12

Lighting Models: 

Lighting Models

Shading Schemes: 

Shading Schemes Flat Shading: same shade to entire polygon

Shading Schemes: 

Shading Schemes Gouraud Shading: smoothly blended intensity across each polygon

Shading Schemes: 

Shading Schemes Phong Shading: interpolated normals to compute intensity at each point Bui Toung Phong Thesis

Scan Convert Polygon P: 

Scan Convert Polygon P

Intensity Interpolation: 

Compute by direction evaluation of illumination expression, whichever formula is being used Intensity Interpolation

Using Average Normals: 

Using Average Normals N = true (geometric) normal

Using Average Normals: 

Using Average Normals

Using Average Normals: 

Using Average Normals

What should corner normals be?: 

What should corner normals be? More generally,

Relevant Light (unit) Vectors: 

Relevant Light (unit) Vectors Point light source direction Surface Normal Reflection direction Viewpoint direction

Computing R Vector: 

Computing R Vector

Computing R Vector: 

Computing R Vector

Computing R Vector: 

Computing R Vector

Flat (Cosine) Shading: 

Flat (Cosine) Shading Compute constant shading function, over each polygon, based on simple cosine term Same normal and light vector across whole polygon Constant shading for polygon

Flat (Cosine) Shading: 

Flat (Cosine) Shading Where, , for unit N, L intensity of point light source diffuse reflection coefficient

Intensity Interpolation (Gouraud): 

Intensity Interpolation (Gouraud)

Slide19: 

Normal Interpolation (Phong)

Slide20: 

Normal Interpolation (Phong) Normalizing makes this a unit vector

Illumination Formula (1/2): 

Illumination Formula (1/2)

Illumination Formula (2/2): 

Illumination Formula (2/2) Where, a denotes ambient term d denotes diffuse term s denotes specular term O denotes object k denotes constant I denotes intensity

Effect of Exponent Parameter: 

Effect of Exponent Parameter As n increases, highlight is more concentrated, surface appears glossier

Mach Band Illusion: 

Mach Band Illusion

Ex: Mach Band Illusion: 

Ex: Mach Band Illusion

Ex: Mach Band Illusion: 

Ex: Mach Band Illusion Mach Bands measured signal perceived signal

Mach Band Illusion: 

Mach Band Illusion How can we avoid the illusion in this image? How do we make the fields appear flat?

Mach Band Illusion: 

Mach Band Illusion Ex: rotating colors eliminate effect

Mach Band Illusion: 

Mach Band Illusion Example (intensity fields are flat!)

Lateral Inhibition: 

Lateral Inhibition Mach Band Illusion is explained by Lateral Inhibition Neighboring visual sensors are connected When one fires, it inhibits its next door neighbors from firing

Lateral Inhibition: 

Lateral Inhibition How does Lateral Inhibition work? Eye sees only difference from surrounding area Eye is a differencing mechanism, a differentiator, an edge detector, a motion detector

Simultaneous Contrast: 

Simultaneous Contrast

Simultaneous Contrast: 

Simultaneous Contrast

Simultaneous Contrast: 

Simultaneous Contrast

Simultaneous Contrast: 

Simultaneous Contrast

Simultaneous Contrast: 

Simultaneous Contrast Georgia O’Keefe, “New York Street with Moon”

Motion Illusion: 

Motion Illusion Illusion Clips of Edward H. Adelson, MIT

Illusions Clips of E. H. Adelson, MIT: 

Illusions Clips of E. H. Adelson, MIT

Illusions Clips of E. H. Adelson, MIT: 

Illusions Clips of E. H. Adelson, MIT

Motion Illusion: 

Motion Illusion Illusion caused by saccadic eye movements

Motion Illusion: 

Motion Illusion

Concentric Circles or Spirals?: 

Concentric Circles or Spirals?

Imposition of Assumptions: 

Imposition of Assumptions

Vasarely: Optical Art: 

Vasarely: Optical Art

Vasarely: Optical Art (2): 

Vasarely: Optical Art (2)

Vasarely: Optical Art (3): 

Vasarely: Optical Art (3)

Vasarely: Optical Art (4): 

Vasarely: Optical Art (4)

Two Categories of Shadows: 

Two Categories of Shadows Two kinds of shadows to consider: study a flag pole The shaded side of the pole is under shadow The pole casts a shadow on the ground.

Two Categories of Shadows: 

Two Categories of Shadows

Progress in Shadows(2): 

Progress in Shadows(2) Eye at light source gives no shadows Separate eye and point source of light Do VSD twice: from light (mark stuff not visible) from eye (shade appropriately)

Progress in Shadows: 

Progress in Shadows Shadow Volumes (Frank Crow) Weiler-Atherton Intersect all polygons againts all others… Big process, not stable, conceptially elegant

Progress in Shadows (3): 

Progress in Shadows (3) Many algorithms, particularly Hechtbert’s “events” General problem: area lights (luminaires) and area blockers Answer can be found in 4D, or higher

Slide53: 

Soft Shadows Complicated structure Seek an analytic solution Point source Area source

Simple Case: One Blocker: 

Simple Case: One Blocker Analytic solution involves splines in 4 dimensions

Slide55: 

Diffuse Scene Direct lighting Indirect lighting

Lighting Effects (1): 

Lighting Effects (1) Multiple Lights Spectral Absorption Texture Surface Irregularities Transparency Refraction

Lighting Effects (2): 

Lighting Effects (2) Atmospheric Effects Subsurface Interaction Shadows Directional Variations (anisotropic) Reflections Environment mapping

Lighting Effects (3): 

Lighting Effects (3) Directional Light Color Bleeding Camera Effects

Ray Tracing: 

Ray Tracing Classical geometric optics technique Extremely versatile Historically viewed as expensive Good for special effects Computationally intensive Can do sophisticated graphics

Ray Tracing: 

Ray Tracing Screen-space point Film plane point in canonical view volume World-space ray

Environment Mapping: 

Environment Mapping

Radiosity: 

Radiosity Considers all light in entire environment Gives indirect lighting Every surface can be an “emitter” Big linear system to solve

Radiosity: 

Radiosity Break up sceen into a system of interacting polygons:

Early Cornell Radiosity Image: 

Early Cornell Radiosity Image Extraordinary image of indirect lighting. Image has color bleeding and warm glow

Early Cornell Radiosity Image: 

Early Cornell Radiosity Image

End Light and Rendering: 

End Light and Rendering 67 Lecture Set 12