A model of visual recognition and categorization
To recognize a previously seen object, the visual system must overcome the variability in the object's appearance caused by factors such as illumination and pose. Developments in computer vision suggest that it may be possible to counter the influence of these factors, by learning to interpolat...
Saved in:
Published in | Philosophical transactions of the Royal Society of London. Series B. Biological sciences Vol. 352; no. 1358; pp. 1191 - 1202 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
England
The Royal Society
29.08.1997
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | To recognize a previously seen object, the visual system must overcome the variability in the object's appearance caused by factors such as illumination and pose. Developments in computer vision suggest that it may be possible to counter the influence of these factors, by learning to interpolate between stored views of the target object, taken under representative combinations of viewing conditions. Daily life situations, however, typically require categorization, rather than recognition, of objects. Due to the open-ended character of both natural and artificial categories, categorization cannot rely on interpolation between stored examples. Nonetheless, knowledge of several representative members, or prototypes, of each of the categories of interest can still provide the necessary computational substrate for the categorization of new instances. The resulting representational scheme based on similarities to prototypes appears to be computationally viable, and is readily mapped onto the mechanisms of biological vision revealed by recent psychophysical and physiological studies. |
---|---|
Bibliography: | istex:2F60022A32F7F97D40E7B022C70EFE63207CCF49 ark:/67375/V84-Z5X4C69H-5 Discussion Meeting Issue 'Knowledge-based vision in man and machine' organized by J. Anderson, H. B. Barlow and R. L. Gregory ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0962-8436 1471-2970 |
DOI: | 10.1098/rstb.1997.0102 |