Key points are not available for this paper at this time.
Reproducing the light field of a 3D scene on a flat-panel display is an ultimate goal of autostereoscopic imaging. Currently, solutions using microlenses or parallax barriers are the most common for this purpose. However, while a light field is a 4D vector space, a flat-panel display is only a 2D entity. There is a fundamental mismatch between the device's capacity and the amount of information to be displayed. To address this problem, we consider a multilayer liquid crystal display (LCD), which is a display device constructed by stacking multiple LCDs on top of a light source. Since the total number of pixels in a multilayer LCD is larger than that of a single layer LCD, a multilayer LCD can store more information than a single layer LCD. Moreover, a multilayer LCD will exhibit varying images depending on the viewers' positions. This property is found to be useful for autostereoscopic 3D viewing, which is elaborated upon further in detail through simulation-based studies.
Hironobu Gotoda (Thu,) studied this question.