Integrating contours within and through depth
Related Experiment Video
Updated: Nov 2, 2026

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
Published on: April 18, 2025
R F Hess1, A Hayes, F A Kingdom
1McGill Vision Research Unit, Department of Ophthalmology, McGill University, Montreal, Quchec, Canada. Rhess@bradman.vision.mcgill.ca
Abstract:
To better understand the role of disparity in contour integration we compared detection performance of "paths" composed of elements confined either to a single depth plane, or spanning multiple depth planes. In both cases paths defined by alignment of elements were embedded in a noise background-field made up of similar, but randomly positioned, elements covering the same depth range as the path elements. We show that a systematic disparity cue can enhance the detectability of paths which traverse depth, but that this detectability is weak compared to paths made up of elements of the same disparity. These results suggest that the outputs of disparity detectors tuned to different disparities can be linked to define contours.
Related Concept Videos
Area Between Curves: Integrating With Respect to x
Area Between Curves: Integrating With Respect to y
Level Curves and Contour Maps
Depth Perception and Spatial Vision
Tangent Planes to Level Surfaces
Topographic Surveying and Contours
