Page 114 -
P. 114

3.1 Point operators                                                                     93











                         (a)                    (b)                    (c)                     (d)

               Figure 3.4 Image matting and compositing (Chuang, Curless, Salesin et al. 2001) c   2001 IEEE: (a) source
               image; (b) extracted foreground object F; (c) alpha matte α shown in grayscale; (d) new composite C.






                                      ×   (1−               )  +                  =



                               B                   α                  αF                  C
                              (a)                 (b)                 (c)                (d)
               Figure 3.5 Compositing equation C =(1 − α)B + αF. The images are taken from a close-up of the region of
               the hair in the upper right part of the lion in Figure 3.4.



               1996), while the process of inserting it into another image (without visible artifacts) is called
               compositing (Porter and Duff 1984; Blinn 1994a).
                  The intermediate representation used for the foreground object between these two stages
               is called an alpha-matted color image (Figure 3.4b–c). In addition to the three color RGB
               channels, an alpha-matted image contains a fourth alpha channel α (or A) that describes the
               relative amount of opacity or fractional coverage at each pixel (Figures 3.4c and 3.5b). The
               opacity is the opposite of the transparency. Pixels within the object are fully opaque (α =1),
               while pixels fully outside the object are transparent (α =0). Pixels on the boundary of the
               object vary smoothly between these two extremes, which hides the perceptual visible jaggies
               that occur if only binary opacities are used.
                  To composite a new (or foreground) image on top of an old (background) image, the over
               operator, first proposed by Porter and Duff (1984) and then studied extensively by Blinn
               (1994a; 1994b), is used,
                                           C =(1 − α)B + αF.                         (3.8)

               This operator attenuates the influence of the background image B by a factor (1 − α) and
               then adds in the color (and opacity) values corresponding to the foreground layer F, as shown
               in Figure 3.5.
                  In many situations, it is convenient to represent the foreground colors in pre-multiplied
               form, i.e., to store (and manipulate) the αF values directly. As Blinn (1994b) shows, the
               pre-multiplied RGBA representation is preferred for several reasons, including the ability
               to blur or resample (e.g., rotate) alpha-matted images without any additional complications
               (just treating each RGBA band independently). However, when matting using local color
               consistency (Ruzon and Tomasi 2000; Chuang, Curless, Salesin et al. 2001), the pure un-
   109   110   111   112   113   114   115   116   117   118   119