This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Image Compositing Image compositing is the blending of two or more images into a single output image, where each pixel is a combination of the colors from the corresponding pixels of the source images. In most cases two images are combined to make a single image. This compositing operation is also often referred to as keying . The most common example of this technique is chromakeying . Chromakeying creates an output image from a foreground and background image. Pixels where the foreground image is some selected color, usually blue or green, are replaced with pixels from the background image. The net effect is that a person can be photographed in front of a blue screen and a background image can be keyed, replacing the blue in the picture with an image the person appears to be standing in front of. Figure 1 is an example of chromakeying. Figure 1 - Chromakey example Alpha Blending The color of the pixels in the output image is determined by blending the colors of the pixels in the input image using a method called alpha blending. Alpha ( α ) is a floating point value that ranges from 0 to 1, where 0 indicates the color of the pixel will be that of the background and 1 indicates the color of the pixel will be that of the foreground. Values between 0 and 1 will indicate a blending of the two colors. This is referred to as alpha blending . A simple linear blend is applied as illustrated in Equation 1: 1 1 ) 1 ( c c c r α α − + = 1 In this equation c r is the color of the output pixel, c 1 is the color of the foreground pixel, and c 2 is the color of the background pixel. α is the blending value, which ranges from 0 to 1. This chapter will use the convention that image 1 is the foreground image and image 2 is the background image. Result images will be labeled with the letter R. The ImageComposite application Figure 2 - ImageComposite application The ImageComposite application is a program supplied with the text that allows for image compositing experimentation. An example screen from ImageComposite is shown in Figure 2. An entry box is provided where an equation that computes an alpha value can be entered into the program. The variables r1, g1, b1, and l1 are the red, green, blue, and luminance values respectively of image 1, the foreground image. The variables r2, g2, b2, and l2 are the values for image 2. These variables are in the range 0 to 1. Because keying is nearly always based only on the colors in the foreground image, equations entered into ImageComposite will usually only have r1, g1, b1, and l1 values in them. Values a1, a2, a3, a4, a5, and a6 can be entered as variable in equations. These variables take on values set by the sliders shown in Figure 2. The sliders can be adjusted dynamically to change the value of the variables and experiment with values in the equations. ImageComposite also supports blue and green spill with a spill adjustment on the sliders. the sliders....
View Full Document
This note was uploaded on 07/25/2008 for the course CSE 471 taught by Professor Owen during the Fall '07 term at Michigan State University.
- Fall '07