Correcting interperspective aliasing in autostereoscopic displays

Abstract

An image presented on an autostereoscopic system should not contain discontinuities between adjacent views. A viewer should experience a continuous scene when moving from one view to the next. If corresponding points in two perspectives do not spatially abut, a viewer will experience jumps in the scene. This is known as interperspective aliasing. Interperspective aliasing is caused by object features far away from the stereoscopic screen being too small, which results in visual artifacts. By modeling a 3D point as a defocused image point, we can adapt Fourier analysis to devise a depth-dependent filter kernel that allows filtering of a stereoscopic 3D image. For synthetic 3D data, we use a simpler approach, which is to smear the data by a distance proportional to its depth.

Resampling, Antialiasing, and Compression in Multiview 3-D Displays

“…He models stereograms as imaging systems with a generalized exit pupil [6] that corresponds in size to the spacing v between the virtual perspective views ( Figure 2). In addition, he includes a focusing error that models the fact that light waves originate at a distance f from the viewer instead of the true distance f + d. St. Hilaire's results are applicable not only to holographic stereograms but also to other techniques to multiplex perspective views, such as parallax barriers and lenticular arrays [18]. His analysis shows that the OTF (and the point spread function) is affected by diffraction effects only if the spacing between the centers of projection of the perspective views ( v in Figure 2) is smaller than about 1 mm at a typical viewing distance of about 60 cm.…”

“…Halle [8] and Moller and Travis [18] derive several antialiasing algorithms from the depth dependent point spread function in (1). One approach is to filter the image plane projection of each point based on its depth.…”

“…One simply convolves the ray-space representation of the input signal with a 2-D display prefilter corresponding to the bandwidth in Figure 3 as proposed by Zwicker et al [26]. This approach has also been described informally by Halle [8] and Moller and Travis [18]. Intuitively, it averages the perspective views that lie between the centers of projection sampled by the display.…”

Display pre-filtering for multi-view video compression

“…Moller et al [15] describe a method to prevent interperspective aliasing that is based on St Hilaire's [6] display bandwidth analysis. Unfortunately, this approach requires the knowledge of per pixel scene depth.…”

“…Therefore, many interesting scenes exceed the display bandwidth. Rendering these scenes leads to inter-perspective aliasing [15].…”