Search PubMedSearch

Biomedical subjects

J H Rieger

Publications and source records attributed to J H Rieger.

3 recordsLinked to original sources

Human visual navigation in the presence of 3-D rotations.

We report on the ability of human observers in judging their direction of translation from sparse, moving random dot patterns for varying extents of 3--D rotation. The observers have to discriminate possible axes of translation with angular separations of 2.5 deg or 5 deg. The field of view is either 20 X 20 deg or 10 X 10 deg. The simulated observer movement is relative to two types of scenes. The first type consists of dots located on a single plane at a depth Z. The second type of scenes consists of dots located on two transparent planes at different depths Z and Z + dZ. Unlike in the single plane condition, where the judgements about the direction of translation deteriorate quickly as the magnitude of 3--D rotation increases, we find for movements relative to planes at different distances a stable performance over a range of rotational magnitudes. Moreover we find that a reduction of the field of view from 20 X 20 deg to 10 X 10 deg does not affect the judgements significantly.

Depth Perception

Processing differential image motion.

The inference of three-dimensional camera motion parameters and the layout of a scene from image flows becomes particularly simple from a computational point of view if the scene contains depth variations. Under this condition, the differential image motion yields a simple estimate of the translation field lines at image locations corresponding to depth discontinuities in the scene. This in turn facilitates closed-form solutions of camera motion parameters and environmental depth. Our results may have relevance to human motion perception, which also seems to rely on depth variation in processing image motion.

Depth Perception

Information in optical flows induced by curved paths of observation.

For a moving observer it is essential to foresee the locomotor course with respect to structures in the environment. Optical flows that are available to a moving observer contain powerful information for visual kinesthesis. In general, optical flows consist of separable translational and rotational components. The information examined here is contained completely in the translational component and its time derivatives. Curved paths of observation are specified by different orientations of the translational components of optical velocity and acceleration fields. Obstacles and their temporal separation from a curvilinearly moving observer are specified in the optical flow, as is the angle of collision.

Humans