Cross-modal slant and curvature matching of stereo- and motion-induced surfaces

Abstract

In many laboratory setups and in many day-to-day situations, a unique solution ofthe structure-fromtwo-views problem is unobtainable. Yet,when the visual system is presented with two projections in a sequence, it nevertheless appears to generate a reasonably stable percept of structure. In the research reported here, we examined whether the same surface would be perceived when subjects were presented with a pair of views that alternated in time monocularly (two-frame motion) or were shown simultaneously to both eyes (stereo). In Experiment 1, we studied slant perception: human observers were asked to match the slant of a motion-induced planar surface with its stereo-induced counterpart. In Experiment 2, the perceived curvature of parabolic surfaces was matched in a similar way. The results show that motion-induced slant is matched with a higher value of the stereo-induced slant. However, the curvature experiment showed that motion-induced curvature is matched with a lower stereo-induced curvature. One possible explanation may be that the slant and curvature are internally inconsistent in at least one of the modalities.Mathematically,the metric three-dimensional (3-D) structure of a rigid set of identifiable points is generally fully specified by only two perspective projections ofthe object from different viewpoints (see, e.g., Longuet-Higgins, 1981).Representations ofsuch 3-D structures can be useful to systems interacting with their environment.The human visual system appears to be able to extract certain (not necessarily Euclidean) 3-D properties of a rigid scene from two distinct projections shown either to both eyes simultaneously (i.e., stereo) or sequentially to one eye (i.e., motion parallax).The displacement of a point in the scene between projections is called disparity in the stereo domain and velocity in the motion domain (given the time frame). The displacement fields in both domains, however, can in principle be processed identically in order to extract 3-D information about the scene. Indeed, ample psychophysical evidence indicates a close relationship between the processing in both modalities Nawrot & Blake, 1989; Norman, Todd, Tittle, & Perotti, 1993; Rogers & Collet, 1989;, 1983.Our aim is to study the ability ofthe visual system to extract and match 3-D properties in these modalities. Our study focuses on the extraction of surface slant and curva-S. C. de V was funded by the SPIN project "3D Computer Vision" of the Dutch Ministry of Economic Affairs; P. w. was funded by the InSight II project of the ESPRIT Basic Research Actions ofthe European Community. The authors would like to thank Harald Vogt for serving as an observer, Maarten Hogervorst and Hendrik-Jan van Veen for many helpful discussions, and Andre Noest for his useful comments on this manuscript in its draft form. Correspondence should be addressed to S. C. de Vries, TNO Human Factors Research Institute, Kampweg 5, 3769 DE Soesterberg, The Netherlands (e-mail: sdevries@fys.ruu.nl or sjoerd@tm.tno.nl).ture specified by either stereo o...

Surface orientation, modulation frequency and the detection and perception of depth defined by binocular disparity and motion parallax

“…This orientation anisotropy is evident at threshold, and also aVects perceived depth magnitude and the time required for depth to be perceived (Bradshaw & Rogers, 1999;Bradshaw, Hibbard, & Gillam, 2002;Hibbard, Bradshaw, Langley, & Rogers, 2002;Mitchison & McKee, 1990;Wallach & Bacon, 1976). Similarly, for a monocular observer making horizontal head movements while viewing a surface shape deWned by the motion parallax arising from these head movements, a similar orientation anisotropy exists which aVects threshold perception and perceived depth magnitude (Allison, Rogers, & Bradshaw, 1978;Cornilleau-Peres & Droulez, 1993;De Vries & Werkhoven, 1995;Rogers & Graham, 1983).…”