Search PubMed⌕ Search

PubMed · 8917773

Non-visual information in structure-from-motion.

Abstract

We examined whether non-visual signals improve visual perception of three-dimensional structure-from-motion. Observers discriminated curvature in quadratic surfaces defined by random dot cinematograms with limited lifetime. They either explored visually a static surface by making head movements that were fed back to the display (HM condition) or they viewed statically the same surface which now rotated (NHM condition). Both conditions showed a clear build-up of performance as lifetime increases, but with different time constants for the HM and NHM condition. A second experiment showed that these differences could not be caused by differences in motion detection for the HM and NHM conditions. We suggest that non-visual information is combined with visual information at a high stage of visual processing, and that it does not mainly serve as input for a retinal stabilization process.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

W J van Damme, W A van de Grind. 1996. Non-visual information in structure-from-motion.. https://doi.org/10.1016/0042-6989(96)00060-0

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

Possibilities and limitations of current stereo-endoscopy.

BACKGROUND: Stereo-endoscopy has become a commonly used technology. In many comparative studies striking advantages of stereo-endoscopy over two-dimensional presentation could not be proven. To show the potential and fields for further improvement of this technology is the aim of this article. METHOD: The physiological basis of three-dimensional vision limitations of current stereo-endoscopes is discussed and fields for further research are indicated. New developments in spatial picture acquisition and spatial picture presentation are discussed. RESULTS: Current limitations of stereo-endoscopy that prevent a better ranking in comparative studies with two-dimensional presentation are mainly based on insufficient picture acquisition. CONCLUSION: Devices for three-dimensional picture presentation are at a more advanced developmental stage than devices for three-dimensional picture acquisition. Further research should emphasize the development of new devices for three-dimensional picture acquisition.

Depth Perception↗

Disparity increment thresholds for gratings.

The classic increment disparity threshold function rises steeply, usually exponentially, with disparity pedestal. Thus a smaller difference in stereoscopic depth can be resolved the nearer it is to the fixation plane. This result has been obtained with relatively broad-bandwidth stimuli. We show here that the increment threshold function for narrow-bandwidth stimuli differs subtly from the classic function: Thresholds vary only modestly over a +/- quarter-cycle pedestal range, by a factor of about 2, and frequently show a dip, yielding best stereo acuity not at the fixation plane but at moderate disparities (20 degrees-30 degrees in phase) on either side of it. Though the dip has not been noted previously, it is consistent with models of disparity processing in which filter sensitivity or selectivity is greatest at a disparity of zero. Moreover, the relatively flat increment threshold function observed at any one scale is compatible with a steeply rising function for broad-bandwidth stimuli.

Depth Perception↗

A computational model of spatio-temporal dynamics in depth filling-in.

We present a computational model based on the heat conduction equation, which can well explain human performance of depth interpolation. The model assumes that the depth information is locally represented and spatial integration is made by iterative processing of mutual interaction of neighbors. It reconstructs a dynamically transforming surface which is in good agreement with the results of psychophysical experiments on depth perception of untextured (uniform-colored) surface moving in depth. The model can also explain a temporal-frequency property of human percept. We conclude that the local ambiguity, which is quite common in everyday visual scenes, is solved by an interpolation mechanism based on iterative local interaction of locally represented visual information.

Depth Perception↗