Conferences Papers

Conferences Papers (39)

Multiobjective design of operators that detect points of interest in images

by
  1. Leonardo Trujillo, Gustavo Olague, Evelyne Lutton and Francisco Fernández Vega. Multiobjective design of operators that detect points of interest in images. In GECCO. 2008, 1299-1306. BibTeX

    @inproceedings{DBLP:conf/gecco/TrujilloOLV08a,
    	author = "Leonardo Trujillo and Gustavo Olague and Evelyne Lutton and Francisco Fern{\'a}ndez de Vega",
    	title = "Multiobjective design of operators that detect points of interest in images",
    	booktitle = "GECCO",
    	year = 2008,
    	pages = "1299-1306",
    	ee = "http://doi.acm.org/10.1145/1389095.1389344",
    	crossref = "DBLP:conf/gecco/2008",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Conor Ryan and Maarten Keijzer (eds.). Genetic and Evolutionary Computation Conference, GECCO 2008, Proceedings, Atlanta, GA, USA, July 12-16, 2008. ACM, 2008. BibTeX

    @proceedings{DBLP:conf/gecco/2008,
    	editor = "Conor Ryan and Maarten Keijzer",
    	title = "Genetic and Evolutionary Computation Conference, GECCO 2008, Proceedings, Atlanta, GA, USA, July 12-16, 2008",
    	booktitle = "GECCO",
    	publisher = "ACM",
    	year = 2008,
    	isbn = "978-1-60558-130-9",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

In this paper, a multiobjective (MO) learning approach to image feature extraction is described, where Pareto-optimal interest point (IP) detectors are synthesized using genetic programming (GP). IPs are image pixels that are unique, robust to changes during image acquisition, and convey highly descriptive information. Detecting such features is ubiquitous to many vision applications, e.g. object recognition, image indexing, stereo vision, and content based image retrieval. In this work, candidate IP operators are automatically synthesized by the GP process using simple image operations and arithmetic functions. Three experimental optimization criteria are considered: 1) the repeatability rate; 2) the amount of global separability between IPs; and 3) the information content captured by the set of detected IPs. The MO-GP search considers Pareto dominance relations between candidate operators, a perspective that has not been contemplated in previous research devoted to this problem. The experimental results suggest that IP detection is an illposed problem for which a single globally optimum solution does not exist. We conclude that the evolved operators outperform and dominate, in the Pareto sense, all previously man-made designs.

Published in
Proceedings of the 10th annual conference on Genetic and evolutionary computation
Pages 1299-1306
http://dl.acm.org/citation.cfm?id=1389095.1389344&coll=ACM&dl=GUIDE&type=series&idx=SERIES11264&part=series&WantType=Proceedings&title=GECCO
Date of conference
12 - 16 July 2008
ISBN
978-1-60558-130-9
Read more...

Behavior-based speciation for evolutionary robotics

by
  1. Leonardo Trujillo, Gustavo Olague, Evelyne Lutton and Francisco Fernández Vega. Behavior-based speciation for evolutionary robotics. In GECCO. 2008, 297-298. BibTeX

    @inproceedings{DBLP:conf/gecco/TrujilloOLV08,
    	author = "Leonardo Trujillo and Gustavo Olague and Evelyne Lutton and Francisco Fern{\'a}ndez de Vega",
    	title = "Behavior-based speciation for evolutionary robotics",
    	booktitle = "GECCO",
    	year = 2008,
    	pages = "297-298",
    	ee = "http://doi.acm.org/10.1145/1389095.1389147",
    	crossref = "DBLP:conf/gecco/2008",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Conor Ryan and Maarten Keijzer (eds.). Genetic and Evolutionary Computation Conference, GECCO 2008, Proceedings, Atlanta, GA, USA, July 12-16, 2008. ACM, 2008. BibTeX

    @proceedings{DBLP:conf/gecco/2008,
    	editor = "Conor Ryan and Maarten Keijzer",
    	title = "Genetic and Evolutionary Computation Conference, GECCO 2008, Proceedings, Atlanta, GA, USA, July 12-16, 2008",
    	booktitle = "GECCO",
    	publisher = "ACM",
    	year = 2008,
    	isbn = "978-1-60558-130-9",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

This paper describes a speciation method that allows an evolutionary process to learn several robot behaviors using a single execution. Species are created in behavioral space in order to promote the discovery of different strategies that can solve the same navigation problem. Candidate neurocontrollers are grouped into species based on their corresponding behavior signature, which represents the traversed path of the robot within the environment.Behavior signatures are encoded using character strings and are compared using the string edit distance. The proposed approach is better suited for an evolutionary robotics problem than speciating in objective or topological space. Experimental comparison with the NEAT method confirms the usefulness of the proposal.

Published in
GECCO '08 Proceedings of the 10th annual conference on Genetic and evolutionary computation
Pages 297-298
http://dl.acm.org/citation.cfm?id=1389147
Date of conference
12 - 18 July 2008
ISBN
978-1-60558-130-9
Read more...

Selecting local region descriptors with a genetic algorithm for real-world place recognition

by
  1. Leonardo Trujillo, Gustavo Olague, Francisco Fernández Vega and Evelyne Lutton. Selecting Local Region Descriptors with a Genetic Algorithm for Real-World Place Recognition. In EvoWorkshops. 2008, 325-334. BibTeX

    @inproceedings{DBLP:conf/evoW/TrujilloOVL08,
    	author = "Leonardo Trujillo and Gustavo Olague and Francisco Fern{\'a}ndez de Vega and Evelyne Lutton",
    	title = "Selecting Local Region Descriptors with a Genetic Algorithm for Real-World Place Recognition",
    	booktitle = "EvoWorkshops",
    	year = 2008,
    	pages = "325-334",
    	ee = "http://dx.doi.org/10.1007/978-3-540-78761-7_33",
    	crossref = "DBLP:conf/evoW/2008",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Mario Giacobini, Anthony Brabazon, Stefano Cagnoni, Gianni Di Caro, Rolf Drechsler, Anikó Ekárt, Anna Esparcia-Alcázar, Muddassar Farooq, Andreas Fink, Jon McCormack, Michael O'Neill, Juan Romero, Franz Rothlauf, Giovanni Squillero, Sima Uyar and Shengxiang Yang (eds.). Applications of Evolutionary Computing, EvoWorkshops 2008: EvoCOMNET, EvoFIN, EvoHOT, EvoIASP, EvoMUSART, EvoNUM, EvoSTOC, and EvoTransLog, Naples, Italy, March 26-28, 2008. Proceedings 4974. Springer, 2008. BibTeX

    @proceedings{DBLP:conf/evoW/2008,
    	editor = "Mario Giacobini and Anthony Brabazon and Stefano Cagnoni and Gianni Di Caro and Rolf Drechsler and Anik{\'o} Ek{\'a}rt and Anna Esparcia-Alc{\'a}zar and Muddassar Farooq and Andreas Fink and Jon McCormack and Michael O'Neill and Juan Romero and Franz Rothlauf and Giovanni Squillero and Sima Uyar and Shengxiang Yang",
    	title = "Applications of Evolutionary Computing, EvoWorkshops 2008: EvoCOMNET, EvoFIN, EvoHOT, EvoIASP, EvoMUSART, EvoNUM, EvoSTOC, and EvoTransLog, Naples, Italy, March 26-28, 2008. Proceedings",
    	booktitle = "EvoWorkshops",
    	publisher = "Springer",
    	series = "Lecture Notes in Computer Science",
    	volume = 4974,
    	year = 2008,
    	isbn = "978-3-540-78760-0",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

The basic problem for a mobile vision system is determining where it is located within the world. In this paper, a recognition system is presented that is capable of identifying known places such as rooms and corridors. The system relies on a bag of features approach using locally prominent image regions. Real-world locations are modeled using a mixture of Gaussians representation, thus allowing for a multimodal scene characterization. Local regions are represented by a set of 108 statistical descriptors computed from different modes of information. From this set the system needs to determine which subset of descriptors captures regularities between image regions of the same location, and also discriminates between regions of different places. A genetic algorithm is used to solve this selection task, using a fitness measure that promotes: 1) a high classification accuracy; 2) the selection of a minimal subset of descriptors; and 3) a high separation among place models. The approach is tested on two real world examples: a) using a sequence of still images with 4 different locations; and b) a sequence that contains 8 different locations. Results confirm the ability of the system to identify previously seen places in a real-world setting.

Published in
10th European Workshop on Evolutionary Computation in Image Analysis and Signal Processing (EvoASP'08)
Volume 4974
Pages 325-334
http://link.springer.com/chapter/10.1007%2F978-3-540-78761-7_33
Date of conference
24 - 26 March 2008
ISSN
0302-9743
ISBN
978-3-540-78761-7

 

Read more...

Discovering several robot behaviors through speciation

by
  1. Leonardo Trujillo, Gustavo Olague, Evelyne Lutton and Francisco Fernández Vega. Discovering Several Robot Behaviors through Speciation. In EvoWorkshops. 2008, 164-174. BibTeX

    @inproceedings{DBLP:conf/evoW/TrujilloOLV08,
    	author = "Leonardo Trujillo and Gustavo Olague and Evelyne Lutton and Francisco Fern{\'a}ndez de Vega",
    	title = "Discovering Several Robot Behaviors through Speciation",
    	booktitle = "EvoWorkshops",
    	year = 2008,
    	pages = "164-174",
    	ee = "http://dx.doi.org/10.1007/978-3-540-78761-7_17",
    	crossref = "DBLP:conf/evoW/2008",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Mario Giacobini, Anthony Brabazon, Stefano Cagnoni, Gianni Di Caro, Rolf Drechsler, Anikó Ekárt, Anna Esparcia-Alcázar, Muddassar Farooq, Andreas Fink, Jon McCormack, Michael O'Neill, Juan Romero, Franz Rothlauf, Giovanni Squillero, Sima Uyar and Shengxiang Yang (eds.). Applications of Evolutionary Computing, EvoWorkshops 2008: EvoCOMNET, EvoFIN, EvoHOT, EvoIASP, EvoMUSART, EvoNUM, EvoSTOC, and EvoTransLog, Naples, Italy, March 26-28, 2008. Proceedings 4974. Springer, 2008. BibTeX

    @proceedings{DBLP:conf/evoW/2008,
    	editor = "Mario Giacobini and Anthony Brabazon and Stefano Cagnoni and Gianni Di Caro and Rolf Drechsler and Anik{\'o} Ek{\'a}rt and Anna Esparcia-Alc{\'a}zar and Muddassar Farooq and Andreas Fink and Jon McCormack and Michael O'Neill and Juan Romero and Franz Rothlauf and Giovanni Squillero and Sima Uyar and Shengxiang Yang",
    	title = "Applications of Evolutionary Computing, EvoWorkshops 2008: EvoCOMNET, EvoFIN, EvoHOT, EvoIASP, EvoMUSART, EvoNUM, EvoSTOC, and EvoTransLog, Naples, Italy, March 26-28, 2008. Proceedings",
    	booktitle = "EvoWorkshops",
    	publisher = "Springer",
    	series = "Lecture Notes in Computer Science",
    	volume = 4974,
    	year = 2008,
    	isbn = "978-3-540-78760-0",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

This contribution studies speciation from the standpoint of evolutionary robotics (ER). A common approach to ER is to design a robot’s control system using neuro-evolution during training. An extension to this methodology is presented here, where speciation is incorporated to the evolution process in order to obtain a varied set of solutions for a robotics problem using a single algorithmic run. Although speciation is common in evolutionary computation, it has been less explored in behavior-based robotics. When employed, speciation usually relies on a distance measure that allows different individuals to be compared. The distance measure is normally computed in objective or phenotypic space. However, the speciation process presented here is intended to produce several distinct robot behaviors; hence, speciation is sought in behavioral space. Thence, individual neurocontrollers are described using behavior signatures, which represent the traversed path of the robot within the training environment and are encoded using a character string. With this representation, behavior signatures are compared using the normalized Levenshtein distance metric (N-GLD). Results indicate that speciation in behavioral space does indeed allow the ER system to obtain several navigation strategies for a common experimental setup. This is illustrated by comparing the best individual from each species with those obtained using the Neuro-Evolution of Augmenting Topologies (NEAT) method which speciates neural networks in topological space.

Published in
4th European Workshop on Bio-Inspired Heuristics for Design Automation (EvoHOT'08)
Volume 4974
Pages 164-174
http://link.springer.com/chapter/10.1007%2F978-3-540-78761-7_17
(Best Paper Award at EvoHOT 2008)
Date of conference
24 - 26 March 2008
ISSN
0302-9743
ISBN
978-3-540-78761-7
Read more...

Customizable execution environments with virtual desktop grid computing

by
  1. D Lombraña González, Fernández F Vega, L Trujillo, G Olague and Ben Segal. Customizable execution environments with virtual desktop grid computing. In Proceedings of the 19th IASTED International Conference on Parallel and Distributed Computing and Systems. 2007, 7–12. URL BibTeX

    @inproceedings{LombranaGonzalez:2007:CEE:1647539.1647542,
    	author = "Lombra\~{n}a Gonz\'{a}lez, D. and de Vega, F. Fern\'{a}ndez and Trujillo, L. and Olague, G. and Segal, Ben",
    	title = "Customizable execution environments with virtual desktop grid computing",
    	booktitle = "Proceedings of the 19th IASTED International Conference on Parallel and Distributed Computing and Systems",
    	series = "PDCS '07",
    	year = 2007,
    	isbn = "978-0-88986-704-8",
    	location = "Cambridge, Massachusetts",
    	pages = "7--12",
    	numpages = 6,
    	url = "http://dl.acm.org/citation.cfm?id=1647539.1647542",
    	acmid = 1647542,
    	publisher = "ACTA Press",
    	address = "Anaheim, CA, USA",
    	keywords = "desktop grid computing, parallel computing, virtual machines, virtualization"
    }
    
Abstract

Nowadays, desktop machines have good features in terms of computing power, but they are still normally underused at research centers (universities, companies, etc.). On the other hand, some researchers cannot solve very complex problems because they lack sufficient computing power. In this paper, we propose to exploit commodity machines by using Desktop Grid Computing (DGC) technology. Moreover, we employ this kind of infrastructure allowing researchers to deploy and run their applications without any code changes. The goal is to provide an on-demand customized execution environment where scientists can load and run their applications and experiments without worrying about the underlying hardware and operating systems of the client desktop machines. Our proposal achieves this objective by using a DGC technology, such as BOINC, in conjunction with a Virtual Machine technology like VMware. In this paper we introduce this new approach to DGC computing and analyze the successful results that we have obtained.

Published in
PDCS '07 Proceedings of the 19th IASTED International Conference on Parallel and Distributed Computing and Systems
http://dl.acm.org/citation.cfm?id=1647542
Date of conference
19 November 2007
Pages
7-12
ISBN
978-0-88986-704-8
Read more...

Scale invariance for evolved interested operators

by
  1. Leonardo Trujillo and Gustavo Olague. Scale Invariance for Evolved Interest Operators. In EvoWorkshops. 2007, 423-430. BibTeX

    @inproceedings{DBLP:conf/evoW/TrujilloO07,
    	author = "Leonardo Trujillo and Gustavo Olague",
    	title = "Scale Invariance for Evolved Interest Operators",
    	booktitle = "EvoWorkshops",
    	year = 2007,
    	pages = "423-430",
    	ee = "http://dx.doi.org/10.1007/978-3-540-71805-5_47",
    	crossref = "DBLP:conf/evoW/2007",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Mario Giacobini, Anthony Brabazon, Stefano Cagnoni, Gianni Di Caro, Rolf Drechsler, Muddassar Farooq, Andreas Fink, Evelyne Lutton, Penousal Machado, Stefan Minner, Michael O'Neill, Juan Romero, Franz Rothlauf, Giovanni Squillero, Hideyuki Takagi, Sima Uyar and Shengxiang Yang (eds.). Applications of Evolutinary Computing, EvoWorkshops 2007: EvoCoMnet, EvoFIN, EvoIASP,EvoINTERACTION, EvoMUSART, EvoSTOC and EvoTransLog, Valencia, Spain, April11-13, 2007, Proceedings 4448. Springer, 2007. BibTeX

    @proceedings{DBLP:conf/evoW/2007,
    	editor = "Mario Giacobini and Anthony Brabazon and Stefano Cagnoni and Gianni Di Caro and Rolf Drechsler and Muddassar Farooq and Andreas Fink and Evelyne Lutton and Penousal Machado and Stefan Minner and Michael O'Neill and Juan Romero and Franz Rothlauf and Giovanni Squillero and Hideyuki Takagi and Sima Uyar and Shengxiang Yang",
    	title = "Applications of Evolutinary Computing, EvoWorkshops 2007: EvoCoMnet, EvoFIN, EvoIASP,EvoINTERACTION, EvoMUSART, EvoSTOC and EvoTransLog, Valencia, Spain, April11-13, 2007, Proceedings",
    	booktitle = "EvoWorkshops",
    	publisher = "Springer",
    	series = "Lecture Notes in Computer Science",
    	volume = 4448,
    	year = 2007,
    	isbn = "978-3-540-71804-8",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

This work presents scale invariant region detectors that apply evolved operators to extract an interest measure. We evaluate operators using their repeatability rate, and have experimentally identified a plateau of local optima within a space of possible interest operators Ω. The space Ω contains operators constructed with Gaussian derivatives and standard arithmetic operations. From this set of local extrema, we have chosen two operators, obtained by searching within Ω using Genetic Programming, that are optimized for high repeatability and global separability when imaging conditions are modified by a known transformation. Then, by embedding the operators into the linear scale space generated with a Gaussian kernel we can characterize scale invariant features by detecting extrema within the scale space response of each operator. Our scale invariant region detectors exhibit a high performance when compared with state-of-the-art techniques on standard tests.

Published in
Applications of Evolutionary Computing
Lecture Notes in Computer Science
Volume 4448
http://link.springer.com/chapter/10.1007%2F978-3-540-71805-5_47
Date of conference
11 - 13 Abril 2007
Pages
423-430
ISSN
0302-9743
ISBN
978-3-540-71805-5
Read more...

Multiclass object recognition based on texture linear genetic programming

by
  1. Gustavo Olague, Eva Romero, Leonardo Trujillo and Bir Bhanu. Multiclass Object Recognition Based on Texture Linear Genetic Programming. In EvoWorkshops. 2007, 291-300. BibTeX

    @inproceedings{DBLP:conf/evoW/OlagueRTB07,
    	author = "Gustavo Olague and Eva Romero and Leonardo Trujillo and Bir Bhanu",
    	title = "Multiclass Object Recognition Based on Texture Linear Genetic Programming",
    	booktitle = "EvoWorkshops",
    	year = 2007,
    	pages = "291-300",
    	ee = "http://dx.doi.org/10.1007/978-3-540-71805-5_32",
    	crossref = "DBLP:conf/evoW/2007",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Mario Giacobini, Anthony Brabazon, Stefano Cagnoni, Gianni Di Caro, Rolf Drechsler, Muddassar Farooq, Andreas Fink, Evelyne Lutton, Penousal Machado, Stefan Minner, Michael O'Neill, Juan Romero, Franz Rothlauf, Giovanni Squillero, Hideyuki Takagi, Sima Uyar and Shengxiang Yang (eds.). Applications of Evolutinary Computing, EvoWorkshops 2007: EvoCoMnet, EvoFIN, EvoIASP,EvoINTERACTION, EvoMUSART, EvoSTOC and EvoTransLog, Valencia, Spain, April11-13, 2007, Proceedings 4448. Springer, 2007. BibTeX

    @proceedings{DBLP:conf/evoW/2007,
    	editor = "Mario Giacobini and Anthony Brabazon and Stefano Cagnoni and Gianni Di Caro and Rolf Drechsler and Muddassar Farooq and Andreas Fink and Evelyne Lutton and Penousal Machado and Stefan Minner and Michael O'Neill and Juan Romero and Franz Rothlauf and Giovanni Squillero and Hideyuki Takagi and Sima Uyar and Shengxiang Yang",
    	title = "Applications of Evolutinary Computing, EvoWorkshops 2007: EvoCoMnet, EvoFIN, EvoIASP,EvoINTERACTION, EvoMUSART, EvoSTOC and EvoTransLog, Valencia, Spain, April11-13, 2007, Proceedings",
    	booktitle = "EvoWorkshops",
    	publisher = "Springer",
    	series = "Lecture Notes in Computer Science",
    	volume = 4448,
    	year = 2007,
    	isbn = "978-3-540-71804-8",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

This paper presents a linear genetic programming approach, that solves simultaneously the region selection and feature extraction tasks, that are applicable to common image recognition problems. The method searches for optimal regions of interest, using texture information as its feature space and classification accuracy as the fitness function. Texture is analyzed based on the gray level cooccurrence matrix and classification is carried out with a SVM committee. Results show effective performance compared with previous results using a standard image database.

Published in
Applications of Evolutionary Computing
Lecture Notes in Computer Science
Volume 4448
http://link.springer.com/chapter/10.1007%2F978-3-540-71805-5_32
(Nominated for the Best Paper Award at EvoIASP 2007)
Date of conference
11 - 13 Abril 2007
Pages
291 - 300
ISBN
0-7695-2372-2
Read more...

Using Evolution to Learn How to Perform Interest Point Detection

by
  1. Leonardo Trujillo and Gustavo Olague. Using Evolution to Learn How to Perform Interest Point Detection. In ICPR (1). 2006, 211-214. BibTeX

    @inproceedings{DBLP:conf/icpr/TrujilloO06,
    	author = "Leonardo Trujillo and Gustavo Olague",
    	title = "Using Evolution to Learn How to Perform Interest Point Detection",
    	booktitle = "ICPR (1)",
    	year = 2006,
    	pages = "211-214",
    	ee = "http://doi.ieeecomputersociety.org/10.1109/ICPR.2006.1153",
    	crossref = "DBLP:conf/icpr/2006",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. 18th International Conference on Pattern Recognition (ICPR 2006), 20-24 August 2006, Hong Kong, China. IEEE Computer Society, 2006. BibTeX

    @proceedings{DBLP:conf/icpr/2006,
    	title = "18th International Conference on Pattern Recognition (ICPR 2006), 20-24 August 2006, Hong Kong, China",
    	booktitle = "ICPR",
    	publisher = "IEEE Computer Society",
    	year = 2006,
    	isbn = "0-7695-2521-0",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

The performance of high-level computer vision applications is tightly coupled with the low-level vision operations that are commonly required. Thus, it is advantageous to have low-level feature extractors that are optimal with respect to a desired performance criteria. This paper presents a novel approach that uses genetic programming as a learning framework that generates a specific type of low-level feature extractor: interest point detector. The learning process is posed as an optimization problem. The optimization criterion is designed to promote the emergence of the detectors' geometric stability under different types of image transformations and global separability between detected points. This concept is represented by the operators repeatability rate. Results prove that our approach is effective at automatically generating low-level feature extractors. This paper presents two different evolved operators: IPGP1 and IPGP2. Their performance is comparable with the Harris operator given their excellent repeatability rate. Furthermore, the learning process was able to rediscover the DET corner detector proposed by Beaudet

Published in
Pattern Recognition, 2006. ICPR 2006. 18th International Conference on  (Volume:1 )
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=11159
Date of conference
20-24 Aug. 2006
Pages
211 - 214
ISSN
1051-4651
ISBN
0-7695-2521-0
Read more...

Synthesis of interest point detectors through genetic programming

by
  1. Leonardo Trujillo and Gustavo Olague. Synthesis of interest point detectors through genetic programming. In GECCO. 2006, 887-894. BibTeX

    @inproceedings{DBLP:conf/gecco/TrujilloO06,
    	author = "Leonardo Trujillo and Gustavo Olague",
    	title = "Synthesis of interest point detectors through genetic programming",
    	booktitle = "GECCO",
    	year = 2006,
    	pages = "887-894",
    	ee = "http://doi.acm.org/10.1145/1143997.1144151",
    	crossref = "DBLP:conf/gecco/2006",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
  2. Mike Cattolico (ed.). Genetic and Evolutionary Computation Conference, GECCO 2006, Proceedings, Seattle, Washington, USA, July 8-12, 2006. ACM, 2006. BibTeX

    @proceedings{DBLP:conf/gecco/2006,
    	editor = "Mike Cattolico",
    	title = "Genetic and Evolutionary Computation Conference, GECCO 2006, Proceedings, Seattle, Washington, USA, July 8-12, 2006",
    	booktitle = "GECCO",
    	publisher = "ACM",
    	year = 2006,
    	isbn = "1-59593-186-4",
    	bibsource = "DBLP, http://dblp.uni-trier.de"
    }
    
Abstract

This contribution presents a novel approach for the automatic generation of a low-level feature extractor that is useful in higher-level computer vision tasks. Specifically, our work centers on the well-known computer vision problem of interest point detection. We pose interest point detection as an optimization problem, and are able to apply Genetic Programming to generate operators that exhibit human-competitive performace when compared with state-of-the-art designs. This work uses the repeatability rate that is applied as a benchmark metric in computer vision literature as part of the GP fitness function, together with a measure of the entropy related with the point distribution across the image. This two measures promote geometric stability and global separability under several types of image transformations. This paper introduces a Genetic Programming implementation that was able to discover a modified version of the DET operator [3], that shows a surprisingly high-level of performace. In this work emphasis was given to the balance between genetic programming and domain knowledge expertise to obtain results that are equal or better than human created solutions.

Published in
Proceeding GECCO '06 Proceedings of the 8th annual conference on Genetic and evolutionary computation
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=10468
(Awarded the bronze medal at the 2006 Human-Competetive Awards)
Date of conference
08 - 12 July 2006
Pages
887-894
ISBN
1-59593-186-4
Read more...

Automatic Feature Localization in Thermal Images for Facial Expression Recognition

by
  1. L Trujillo, G Olague, R Hammoud and B Hernandez. Automatic Feature Localization in Thermal Images for Facial Expression Recognition. In Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society Conference on. 2005, 14-14. DOI BibTeX

    @inproceedings{1565309,
    	author = "Trujillo, L. and Olague, G. and Hammoud, R. and Hernandez, B.",
    	booktitle = "Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society Conference on",
    	title = "Automatic Feature Localization in Thermal Images for Facial Expression Recognition",
    	year = 2005,
    	pages = "14-14",
    	keywords = "Data mining;Face detection;Face recognition;Facial features;Feature extraction;Humans;Image analysis;Image recognition;Information analysis;Support vector machines",
    	doi = "10.1109/CVPR.2005.415",
    	issn = "1063-6919"
    }
    
Abstract

We propose an unsupervised Local and Global feature extraction paradigm to approach the problem of facial expression recognition in thermal images. Starting from local, low-level features computed at interest point locations, our approach combines the localization of facial features with the holistic approach. The detailed steps are as follows: First, face localization using bi-modal thresholding is accomplished in order to localize facial features by way of a novel interest point detection and clustering approach. Second, we compute representative Eigenfeatures for feature extraction. Third, facial expression classification is made with a Support Vector Machine Committiee. Finally, the experiments over the IRIS data-set show that automation was achieved with good feature localization and classification performance.

Published in
Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society Conference on
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=10468
Date of conference
25 June 2005
Pages
14
ISSN
1063-6919
ISBN
0-7695-2372-2
Read more...
Subscribe to this RSS feed