Image Compositing. Key Concepts. Colbert Challenge

Transcription

1 CS148: Introduction to Computer Graphics and Imaging Image Compositing Colbert Challenge Key Concepts Optical compositing and mattes The alpha channel Compositing operators Premultipled alpha Matte extraction Page 1

2 Optical Compositing Image Composition Defn: Combine foreground element with background Examples: Graphics arts: Selections and masks, stencils, friskets and masking tape Animation: cels, multiplane camera Film: optical printing, blue screen matting Video: chroma-keying Computer graphics: alpha channel Page 2

3 Animation Cels Appraised on Antiques Roadshow Multiplane Camera Walt Disney Page 3

4 Optical Printing From: Industrial Light and Magic, Thomas Smith (p. 181) From: Special Optical Effects, Zoran Perisic Composing Two Elements + * = Background Holdout Matte + * = Foreground Traveling Matte Page 4

5 The Alpha Channel The Alpha Channel A alpha channel is an additional image that defines: The transparency or opacity of an image The presence or absence of imagery Geometric coverage: soft-edge Or both coverage and transparency Alpha channels may be Binary masks: all or none Mattes: 0.0 to 1.0 (0 to 255) Page 5

6 Fragment: Color + Coverage Pixel C=(R,G,B) = A = Coverage = Area = Opacity = 1 - Transparency Color c of pixel is an area-weighted average of C c = C Area-weighted color = pre-multiplied (with A) color Convention: capital, not multiplied; lower, multiplied Image Composition Page 6

7 OVER Operator F C F B C B A F = F A B = (1- F ) B OVER = Foreground Background Composite Composite color: c = A F C F + A B C B = ( F C F )+ (1- F ) ( B C B ) Composite alpha: = A F + A B = F + (1- F ) B OVER Operator F C F B C B A F = F A B = (1- F ) B OVER = Foreground Background Composite Composite color: c = c F + (1-a F )c B Composite alpha: = F + (1- F ) B Page 7

8 Region Coverage A B + = 4 Regions Porter-Duff Compositing Algebra How many ways can two pixels be combined? Region 1: 1 possibility - 0 Region 2: 2 possibilities - A or 0 Region 3: 2 possibilities - B or 0 Region 4: 3 possibilities - A, B or 0 4 Regions Operators: 12 total possibilities Page 8

9 Porter-Duff Compositing Algebra c = F A c A +F B c B OpenGL Blending Modes Specify src (foreground) and dst (background) F s Porter-Duff 0, 1, As, Ad, 1-As, 1-Ad, Other useful modes min(as,1-ad) Why? Page 9

10 Assumptions Uncorrelated Correlated Anticorrelated Premultiplied Alpha Represent as c = C = ( r, g, b, ) Advantages Closure Recovering C from c would require divide by Display c ; c over K = c + (1- C ) K = c One formula for compositing color and alpha c = c F + (1- F ) c B Less arithmetic Associated: OVER (1 sub, 4 muls, 4adds) Unassociated: OVER (1 sub, 7 muls, 4 adds) Page 10

11 Interpolating RGBA Two ways of interpolating an image 1. Interpolate a and C; then multiply c = a*c 2. Interpolate premultiplied colors c These lead to different answers For example, suppose we interpolate (C, a) = (1,0,0,0) and (0,1,0,1) at ½. 1. (1/2(1+0),1/2(0+1),1/2(0+0),1/2(0+1)) = (1/2,1/2,0,1/2); then we multiply by a to get (1/4,1/4,0,1/2) 2. First we multiply to get (0,0,0,0) and (0,1,0,1); then we interpolate to get (0,1/2,0,1/2) Which Way is Correct? Interpolating composited images Composite foreground over the background and then interpolate Interpolate foreground and background and then composite These should yield the same result! Only true if we interpolate c (method 2 on prev. slide) Page 11

12 Matte Extraction Blue Screen Page 12

13 Pulling a Matte - Matte Creation From digitized images Image processing Set of colors marked transparent, region growing Demonstration: Photoshop Magic Wand Video or chroma-keying Range of luminances marked transparent Blue-screen matting (Petro Vlahos) Separate blue background from foreground image Blue/Green-Screen Matte Extraction Given: C - Observed color (RGB) C B - Backing color = blue or green (RGB) Compute foreground color and alpha (RGBA) c F = ( F R F, F G F, F B F, F ) Page 13

14 Blue/Green-Screen Matte Extraction Compositing equation: C = c F + (1- F ) C B R = ( F R F ) + (1- F ) R B G = ( F G F ) + (1- F ) G B B = ( F B F ) + (1- F ) B B Knowns: (R,G,B) and (R B, G B, B B ) Unknowns: (R F, G F, B F, F ) or (r F, g F, b F, F ) 3 equations in 4 unknowns Petros Vlahos Algorithm Divide color space into two regions The separating plane is given by the equation G = k 2 B That is, colors with lots of green are outside G C B C Outside G = k 2 B K C Inside B Page 14

15 Petros Vlahos Algorithm Observed colors may be inside the region or outside Colors inside are opaque Colors outside are composited with Cb G C B C Outside G = k 2 B K C Inside B Petros Vlahos Algorithm Outside colors c F are on a line between C B and C C = F C F + (1- F ) C B = F + (1- F ) C B => c F = C - (1- F ) C B But where? We don t know F G C B 1 F C F C F K B Page 15

16 Petros Vlahos Algorithm Outside colors are assumed to be on the plane N c F = N C - (1- F ) A C B = 0 We now have a 4 th equation and can solve for 4 unks. G C B N = (0,1,-k 2 ) 1 F C F C F G = k 2 B K B Things to Remember Classic techniques: masks, mattes, optical printing Definition of the alpha channel as opacity/coverage Premultiplied alpha Porter-Duff image compositing algebra Vlahos matte extraction algorithm Page 16