Visual Experience of 3D TV

The goal of this chapter is to explore and determine which evaluation criterion is most appropriate to assess the performance of 3D-display systems. It is assumed that these evaluation criteria take into account image quality as well as reproduced depth. The criterion that weighs depth most in addition to image quality is considered most appropriate. Experiment 1 explores the assessment criteria image quality, depth, naturalness, presence and viewing experience. It presents empirical work on these assessment criteria for stereoscopic image material when using different 2D to 3D conversion algorithms. Results show that viewing experience and naturalness have the most discriminating power between the various algorithms. Hence, the second experiment focuses on these criteria and uses ’perfect’ 3D content with no conversion or depth artefacts. Several noise levels were added to the 2D and 3D images to degrade image quality. The goal is to investigate whether viewing experience or naturalness incorporates depth the most in the absence of depth artefacts. Results show that the noise distortion is weighted equally both with viewing experience and naturalness. Naturalness is more sensitive to depth than viewing experience. This chapter is based on Seuntiëns et al. (2005a) and Lambooij et al. (2005)