A proposal of adaptive probabilistic model of context applied to visual recognition

Billy Peralta, Norman Vergaray, Luis Caro

Resultado de la investigación: Conference contribution

Resumen

Within the increasing automation of tasks, it is necessary to obtain relevant information from images, to have clarity of the actions that must be performed. In this process, it is possible to detect objects individually, however this mode can generate a considerable error due to the enormous number of ways in which an object can be presented. It is therefore necessary to improve the level of accuracy, and a way to achieve this is taking into account the relationships between objects as well. Until now, context-oriented models generate relationships between objects without discriminating between input images. It is necessary that the model operates on each particular image, to reduce the error and obtain a conclusive result. In response to the above, this work proposes an adaptive probabilistic context-model. The model in question is a functional that processes the occurrence probability of the objects in each image, and generates the relations between the detectable objects using a Bayesian network in the form of a tree. This model is updated with the features of each image, requiring a minimization of the quadratic error through a numerical approximation, obtained by Newton-Raphson method. Comparisons were made between different heuristic proposals, as well as tests on different context-oriented databases, in order to validate the results. On the other hand, tests were performed using features extracted through histograms of orient gradients and Deep Learning. It was found that an improvement in the prediction of the order of 20% is feasible in the testing process.

Idioma originalEnglish
Título de la publicación alojada2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018
EditorialIEEE Computer Society
ISBN (versión digital)9781538692332
DOI
EstadoPublished - 2 may 2019
Evento37th International Conference of the Chilean Computer Science Society, SCCC 2018 - Santiago, Chile
Duración: 5 nov 20189 nov 2018

Serie de la publicación

NombreProceedings - International Conference of the Chilean Computer Science Society, SCCC
Volumen2018-November
ISSN (versión impresa)1522-4902

Conference

Conference37th International Conference of the Chilean Computer Science Society, SCCC 2018
PaísChile
CiudadSantiago
Período5/11/189/11/18

Huella dactilar

Bayesian networks
Newton-Raphson method
Automation
Statistical Models
Testing
Deep learning

ASJC Scopus subject areas

  • Engineering(all)
  • Computer Science(all)

Citar esto

Peralta, B., Vergaray, N., & Caro, L. (2019). A proposal of adaptive probabilistic model of context applied to visual recognition. En 2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018 [8705163] (Proceedings - International Conference of the Chilean Computer Science Society, SCCC; Vol. 2018-November). IEEE Computer Society. https://doi.org/10.1109/SCCC.2018.8705163
Peralta, Billy ; Vergaray, Norman ; Caro, Luis. / A proposal of adaptive probabilistic model of context applied to visual recognition. 2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018. IEEE Computer Society, 2019. (Proceedings - International Conference of the Chilean Computer Science Society, SCCC).
@inproceedings{91cd2b77a83543f680c41452780abec8,
title = "A proposal of adaptive probabilistic model of context applied to visual recognition",
abstract = "Within the increasing automation of tasks, it is necessary to obtain relevant information from images, to have clarity of the actions that must be performed. In this process, it is possible to detect objects individually, however this mode can generate a considerable error due to the enormous number of ways in which an object can be presented. It is therefore necessary to improve the level of accuracy, and a way to achieve this is taking into account the relationships between objects as well. Until now, context-oriented models generate relationships between objects without discriminating between input images. It is necessary that the model operates on each particular image, to reduce the error and obtain a conclusive result. In response to the above, this work proposes an adaptive probabilistic context-model. The model in question is a functional that processes the occurrence probability of the objects in each image, and generates the relations between the detectable objects using a Bayesian network in the form of a tree. This model is updated with the features of each image, requiring a minimization of the quadratic error through a numerical approximation, obtained by Newton-Raphson method. Comparisons were made between different heuristic proposals, as well as tests on different context-oriented databases, in order to validate the results. On the other hand, tests were performed using features extracted through histograms of orient gradients and Deep Learning. It was found that an improvement in the prediction of the order of 20{\%} is feasible in the testing process.",
keywords = "Bayesian networks, Functional optimization, Object recognition",
author = "Billy Peralta and Norman Vergaray and Luis Caro",
year = "2019",
month = "5",
day = "2",
doi = "10.1109/SCCC.2018.8705163",
language = "English",
series = "Proceedings - International Conference of the Chilean Computer Science Society, SCCC",
publisher = "IEEE Computer Society",
booktitle = "2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018",
address = "United States",

}

Peralta, B, Vergaray, N & Caro, L 2019, A proposal of adaptive probabilistic model of context applied to visual recognition. En 2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018., 8705163, Proceedings - International Conference of the Chilean Computer Science Society, SCCC, vol. 2018-November, IEEE Computer Society, 37th International Conference of the Chilean Computer Science Society, SCCC 2018, Santiago, Chile, 5/11/18. https://doi.org/10.1109/SCCC.2018.8705163

A proposal of adaptive probabilistic model of context applied to visual recognition. / Peralta, Billy; Vergaray, Norman; Caro, Luis.

2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018. IEEE Computer Society, 2019. 8705163 (Proceedings - International Conference of the Chilean Computer Science Society, SCCC; Vol. 2018-November).

Resultado de la investigación: Conference contribution

TY - GEN

T1 - A proposal of adaptive probabilistic model of context applied to visual recognition

AU - Peralta, Billy

AU - Vergaray, Norman

AU - Caro, Luis

PY - 2019/5/2

Y1 - 2019/5/2

N2 - Within the increasing automation of tasks, it is necessary to obtain relevant information from images, to have clarity of the actions that must be performed. In this process, it is possible to detect objects individually, however this mode can generate a considerable error due to the enormous number of ways in which an object can be presented. It is therefore necessary to improve the level of accuracy, and a way to achieve this is taking into account the relationships between objects as well. Until now, context-oriented models generate relationships between objects without discriminating between input images. It is necessary that the model operates on each particular image, to reduce the error and obtain a conclusive result. In response to the above, this work proposes an adaptive probabilistic context-model. The model in question is a functional that processes the occurrence probability of the objects in each image, and generates the relations between the detectable objects using a Bayesian network in the form of a tree. This model is updated with the features of each image, requiring a minimization of the quadratic error through a numerical approximation, obtained by Newton-Raphson method. Comparisons were made between different heuristic proposals, as well as tests on different context-oriented databases, in order to validate the results. On the other hand, tests were performed using features extracted through histograms of orient gradients and Deep Learning. It was found that an improvement in the prediction of the order of 20% is feasible in the testing process.

AB - Within the increasing automation of tasks, it is necessary to obtain relevant information from images, to have clarity of the actions that must be performed. In this process, it is possible to detect objects individually, however this mode can generate a considerable error due to the enormous number of ways in which an object can be presented. It is therefore necessary to improve the level of accuracy, and a way to achieve this is taking into account the relationships between objects as well. Until now, context-oriented models generate relationships between objects without discriminating between input images. It is necessary that the model operates on each particular image, to reduce the error and obtain a conclusive result. In response to the above, this work proposes an adaptive probabilistic context-model. The model in question is a functional that processes the occurrence probability of the objects in each image, and generates the relations between the detectable objects using a Bayesian network in the form of a tree. This model is updated with the features of each image, requiring a minimization of the quadratic error through a numerical approximation, obtained by Newton-Raphson method. Comparisons were made between different heuristic proposals, as well as tests on different context-oriented databases, in order to validate the results. On the other hand, tests were performed using features extracted through histograms of orient gradients and Deep Learning. It was found that an improvement in the prediction of the order of 20% is feasible in the testing process.

KW - Bayesian networks

KW - Functional optimization

KW - Object recognition

UR - http://www.scopus.com/inward/record.url?scp=85065742046&partnerID=8YFLogxK

U2 - 10.1109/SCCC.2018.8705163

DO - 10.1109/SCCC.2018.8705163

M3 - Conference contribution

T3 - Proceedings - International Conference of the Chilean Computer Science Society, SCCC

BT - 2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018

PB - IEEE Computer Society

ER -

Peralta B, Vergaray N, Caro L. A proposal of adaptive probabilistic model of context applied to visual recognition. En 2018 37th International Conference of the Chilean Computer Science Society, SCCC 2018. IEEE Computer Society. 2019. 8705163. (Proceedings - International Conference of the Chilean Computer Science Society, SCCC). https://doi.org/10.1109/SCCC.2018.8705163