Fixation locations during three-dimensional object recognition are predicted by image segmentation points at concave surface intersections
Keyword(s):