DocumentCode :
1487806
Title :
A maximum-likelihood strategy for directing attention during visual search
Author :
Tagare, Hemant D. ; Toyama, Kentaro ; Wang, Jonathan G.
Author_Institution :
Dept. of Diagnostic Radiol., Dept. of Electr. Eng., Yale Univ., New Haven, CT, USA
Volume :
23
Issue :
5
fYear :
2001
fDate :
5/1/2001 12:00:00 AM
Firstpage :
490
Lastpage :
500
Abstract :
A precise analysis of an entire image is computationally wasteful if one is interested in finding a target object located in a subregion of the image. A useful “attention strategy” can reduce the overall computation by carrying out fast but approximate image measurements and using their results to suggest a promising subregion. The paper proposes a maximum-likelihood attention mechanism that does this. The attention mechanism recognizes that objects are made of parts and that parts have different features. It works by proposing object part and image feature pairings which have the highest likelihood of coming from the target. The exact calculation of the likelihood as well as approximations are provided. The attention mechanism is adaptive, that is, its behavior adapts to the statistics of the image features. Experimental results suggest that, on average, the attention mechanism evaluates less than 2 percent of all part-feature pairs before selecting the actual object, showing a significant reduction in the complexity of visual search
Keywords :
Poisson distribution; object recognition; probability; attention directing; attention strategy; maximum-likelihood attention mechanism; maximum-likelihood strategy; part-feature pairs; visual search; Computer vision; Image analysis; Image converters; Image edge detection; Image recognition; Object recognition; Performance evaluation; Pixel; Statistics; Target recognition;
fLanguage :
English
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
Publisher :
ieee
ISSN :
0162-8828
Type :
jour
DOI :
10.1109/34.922707
Filename :
922707
Link To Document :
بازگشت