Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/17857
Title: Learning to recognize faces incrementally
Authors: Déniz Suárez, Oscar
Lorenzo, J. 
Castrillon, M. 
Mendez, J. 
Falcón Martel, Antonio
UNESCO Clasification: 120304 Inteligencia artificial
Issue Date: 2007
Journal: Lecture Notes in Computer Science 
Conference: 29th Annual Symposium of the German-Association-for-Pattern-Recognition 
29th Annual Symposium of the Deutsche Arbeitsgemeinschaft fur Mustererkennung, DAGM 2007 
Abstract: Most face recognition systems are based on some form of batch learning. Online face recognition is not only more practical, it is also much more biologically plausible. Typical batch learners aim at minimizing both training error and (a measure of) hypothesis complexity. We show that the same minimization can be done incrementally as long as some form of ”scaffolding” is applied throughout the learning process. Scaffolding means: make the system learn from samples that are neither too easy nor too difficult at each step. We note that such learning behavior is also biologically plausible. Experiments using large sequences of facial images support the theoretical claims. The proposed method compares well with other, numerical calculus-based online learners.
URI: http://hdl.handle.net/10553/17857
ISBN: 978-3-540-74933-2
3540749330
ISSN: 0302-9743
DOI: 10.1007/978-3-540-74936-3_37
Source: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)[ISSN 0302-9743],v. 4713 LNCS, p. 365-374
Appears in Collections:Actas de congresos
Thumbnail
Postprint
Adobe PDF (161,19 kB)
Show full item record

SCOPUSTM   
Citations

1
checked on Apr 11, 2020

WEB OF SCIENCETM
Citations

1
checked on May 23, 2020

Page view(s)

10
checked on May 23, 2020

Download(s)

19
checked on May 23, 2020

Google ScholarTM

Check

Altmetric


Share



Export metadata



This item is licensed under a Creative Commons License Creative Commons