Three-dimensional object recognition from single two-dimensional images
Artificial Intelligence
VLP: a visual logic programming language
Journal of Visual Languages and Computing
| Hi-index | 0.00 |
Pictures are inherently ambiguous but people categorize and name pictured objects with remarkable consistency. However, the time to assign a name to a picture depends on a large number of variables, ranging from the quality of the picture itself to the level of hierarchy and frequency of the name. We review the empirical results in the psychological literature on how people name pictured objects, summarizing the major variables that affect the name assigned and the time spent assigning it. The underlying regularities in these data are explained by properties of three mechanisms used in picture naming: bottom-up perceptual encoding; hierarchical associative memory; and top-down knowledge-based search. The properties ascribed to these mechanisms are hypothesized on the basis of computational analyses and considerations of characteristics of the neural systems underlying vision.