The record consists of two parts, Data and Algorithm Demo. The Data component consists of pre-computed Keplerian arrays of all possible matches between filtered random-dot image pairs containing stereoscopically defined surfaces. The Algorithm Demo allows data files to be computed afresh from supplied pairs of images of various surface configurations. Plotting of data is possible in both cases. An interactive demo can also be used to explore the target image selection process.

]]>To implement this strategy, left and right images were convolved with Gaussian kernels of various standard deviations (spatial frequencies). Keplerian arrays comparing filter responses across left and right spatial-frequency combinations were then constructed. Responses that are minimally different across the eyes give rise to regions of high symmetry; position within the Keplerian array indicates the location of a solution in space. Solutions that possess natural surface regularities consistently showed minimal differences for one left : right spatial frequency ratio, which is correlated with local surface slant. As a result, combining responses within particular ratio families can distinguish true matches from false ones. True matches tend to be long and smoothly contoured, and symmetry would be preserved across all members of a ratio family from low to high spatial-frequency combinations.

This approach is efficient; preprocessing is minimal since no feature extraction is involved. It can be implemented in machine vision to solve the correspondence problem for depth sensing algorithms. It is robust when tested against perfectly camouflaged surfaces in random dot stereograms and consistent with physiological data showing that false match signals are propagated to higher cortical areas along the dorsal pathway.

]]>