Quality is assessed by objective and by subjective criteria. Objective criteria are centered around disparity, depth, luminance, contrast, gray shades, and values of the color components such as location in the chromaticity diagram or the noise level in an image. Subjective criteria are harder to define but are subsumed under the perception of structural similarities or the reality of a depth perception. This chapter discusses retrieving quality data, mainly disparity and depth, from given images; to similarities as subjective quality measures; to algorithms for establishing depth maps for 2D and 3D images based on objective and subjective quality measures; and to the reconstruction of 3D images from depth maps and 2D images. The chapter looks at the early, but very instructive, approaches for depth image-based rendering (DIBR) and the state of the art of DIBR is also presented.

Controlled Vocabulary Terms

rendering (computer graphics); three-dimensional displays