Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields

Ognjen Rudovic, Vladimir Pavlovic, Maja Pantic

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

24 Citations (Scopus)
10 Downloads (Pure)

Abstract

Automatic pain intensity estimation from facial images is challenging mainly because of high variability in subject-specific pain expressiveness. This heterogeneity in the subjects causes their facial appearance to vary significantly when experiencing the same pain level. The standard classification methods (e.g., SVMs) do not provide a principled way of accounting for this heterogeneity. To this end, we propose the heteroscedastic Conditional Ordinal Random Field (CORF) model for automatic estimation of pain intensity. This model generalizes the CORF framework for modeling sequences of ordinal variables, by adapting it for heteroscedasticity. This is attained by allowing the variance in the ordinal probit model in the CORF to change depending on the input features, resulting in the model able to adapt to the pain expressiveness level specific to each subject. Our experimental results on the UNBC Shoulder Pain Database show that modeling heterogeneity in the subjects with the framework of CORFs improves the pain intensity estimation attained by the standard CORF model, and the other commonly used classification models.
Original languageEnglish
Title of host publicationAdvances in Visual Computing
Subtitle of host publication9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings
EditorsSabine Coquillart, Xun Luo, Min Chen, David Gotz
Place of PublicationBerlin
PublisherSpringer
Pages234-243
Number of pages10
ISBN (Print)978-3-642-41938-6
DOIs
Publication statusPublished - Jul 2013
Event9th International Symposium on Visual Computing, ISVC 2013 - Rethymnon, Crete, Greece
Duration: 29 Jul 201331 Jul 2013
Conference number: 9

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Verlag
Volume8034
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th International Symposium on Visual Computing, ISVC 2013
Abbreviated titleISVC
CountryGreece
CityRethymnon, Crete
Period29/07/1331/07/13

Keywords

  • EWI-24340
  • METIS-302660
  • IR-89371
  • HMI-HF: Human Factors

Cite this

Rudovic, O., Pavlovic, V., & Pantic, M. (2013). Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields. In S. Coquillart, X. Luo, M. Chen, & D. Gotz (Eds.), Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings (pp. 234-243). (Lecture Notes in Computer Science; Vol. 8034). Berlin: Springer. https://doi.org/10.1007/978-3-642-41939-3_23
Rudovic, Ognjen ; Pavlovic, Vladimir ; Pantic, Maja. / Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields. Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings. editor / Sabine Coquillart ; Xun Luo ; Min Chen ; David Gotz. Berlin : Springer, 2013. pp. 234-243 (Lecture Notes in Computer Science).
@inproceedings{29a2adec3e1e4b29ba1ecace72c6b90d,
title = "Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields",
abstract = "Automatic pain intensity estimation from facial images is challenging mainly because of high variability in subject-specific pain expressiveness. This heterogeneity in the subjects causes their facial appearance to vary significantly when experiencing the same pain level. The standard classification methods (e.g., SVMs) do not provide a principled way of accounting for this heterogeneity. To this end, we propose the heteroscedastic Conditional Ordinal Random Field (CORF) model for automatic estimation of pain intensity. This model generalizes the CORF framework for modeling sequences of ordinal variables, by adapting it for heteroscedasticity. This is attained by allowing the variance in the ordinal probit model in the CORF to change depending on the input features, resulting in the model able to adapt to the pain expressiveness level specific to each subject. Our experimental results on the UNBC Shoulder Pain Database show that modeling heterogeneity in the subjects with the framework of CORFs improves the pain intensity estimation attained by the standard CORF model, and the other commonly used classification models.",
keywords = "EWI-24340, METIS-302660, IR-89371, HMI-HF: Human Factors",
author = "Ognjen Rudovic and Vladimir Pavlovic and Maja Pantic",
year = "2013",
month = "7",
doi = "10.1007/978-3-642-41939-3_23",
language = "English",
isbn = "978-3-642-41938-6",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "234--243",
editor = "Sabine Coquillart and Xun Luo and Min Chen and David Gotz",
booktitle = "Advances in Visual Computing",

}

Rudovic, O, Pavlovic, V & Pantic, M 2013, Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields. in S Coquillart, X Luo, M Chen & D Gotz (eds), Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings. Lecture Notes in Computer Science, vol. 8034, Springer, Berlin, pp. 234-243, 9th International Symposium on Visual Computing, ISVC 2013, Rethymnon, Crete, Greece, 29/07/13. https://doi.org/10.1007/978-3-642-41939-3_23

Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields. / Rudovic, Ognjen; Pavlovic, Vladimir; Pantic, Maja.

Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings. ed. / Sabine Coquillart; Xun Luo; Min Chen; David Gotz. Berlin : Springer, 2013. p. 234-243 (Lecture Notes in Computer Science; Vol. 8034).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields

AU - Rudovic, Ognjen

AU - Pavlovic, Vladimir

AU - Pantic, Maja

PY - 2013/7

Y1 - 2013/7

N2 - Automatic pain intensity estimation from facial images is challenging mainly because of high variability in subject-specific pain expressiveness. This heterogeneity in the subjects causes their facial appearance to vary significantly when experiencing the same pain level. The standard classification methods (e.g., SVMs) do not provide a principled way of accounting for this heterogeneity. To this end, we propose the heteroscedastic Conditional Ordinal Random Field (CORF) model for automatic estimation of pain intensity. This model generalizes the CORF framework for modeling sequences of ordinal variables, by adapting it for heteroscedasticity. This is attained by allowing the variance in the ordinal probit model in the CORF to change depending on the input features, resulting in the model able to adapt to the pain expressiveness level specific to each subject. Our experimental results on the UNBC Shoulder Pain Database show that modeling heterogeneity in the subjects with the framework of CORFs improves the pain intensity estimation attained by the standard CORF model, and the other commonly used classification models.

AB - Automatic pain intensity estimation from facial images is challenging mainly because of high variability in subject-specific pain expressiveness. This heterogeneity in the subjects causes their facial appearance to vary significantly when experiencing the same pain level. The standard classification methods (e.g., SVMs) do not provide a principled way of accounting for this heterogeneity. To this end, we propose the heteroscedastic Conditional Ordinal Random Field (CORF) model for automatic estimation of pain intensity. This model generalizes the CORF framework for modeling sequences of ordinal variables, by adapting it for heteroscedasticity. This is attained by allowing the variance in the ordinal probit model in the CORF to change depending on the input features, resulting in the model able to adapt to the pain expressiveness level specific to each subject. Our experimental results on the UNBC Shoulder Pain Database show that modeling heterogeneity in the subjects with the framework of CORFs improves the pain intensity estimation attained by the standard CORF model, and the other commonly used classification models.

KW - EWI-24340

KW - METIS-302660

KW - IR-89371

KW - HMI-HF: Human Factors

U2 - 10.1007/978-3-642-41939-3_23

DO - 10.1007/978-3-642-41939-3_23

M3 - Conference contribution

SN - 978-3-642-41938-6

T3 - Lecture Notes in Computer Science

SP - 234

EP - 243

BT - Advances in Visual Computing

A2 - Coquillart, Sabine

A2 - Luo, Xun

A2 - Chen, Min

A2 - Gotz, David

PB - Springer

CY - Berlin

ER -

Rudovic O, Pavlovic V, Pantic M. Automatic Pain Intensity Estimation using Heteroscedastic Conditional Ordinal Random Fields. In Coquillart S, Luo X, Chen M, Gotz D, editors, Advances in Visual Computing: 9th International Symposium, ISVC 2013, Rethymnon, Crete, Greece, July 29-31, 2013. Proceedings. Berlin: Springer. 2013. p. 234-243. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-642-41939-3_23