Temporal alignment improves feature quality: An experiment on activity recognition with accelerometer data

Hongjun Choi, Qiao Wang, Meynard Toledo, Pavan Turaga, Matthew Buman, Anuj Srivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Activity recognition has been receiving significant attention from a variety of research areas such as human performance enhancement, health promotion, and human computer interaction. However, recognizing activities from accelerometer data still remains a challenging problem due to sensitivity to sampling rates, misalignment of data, and increased variability in activities among clinically relevant populations. In order to solve these issues, we adopt methods from functional analysis, which consider non-elastic rate variations in movement. The overall framework factors out temporal variability within activity classes, before leveraging robust machine learning pipelines for a given end-use. The proposed approach has been evaluated on 7 classes of everyday activities with 50 subjects. The results indicate that proposed approach achieves improved performance with the improvements observed in separating similar classes that differ in temporal rates, and also demonstrate higher robustness to change in window lengths. These results suggest that temporal alignment should be considered a core part of activity recognition pipelines.

Original languageEnglish (US)
Title of host publicationProceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018
PublisherIEEE Computer Society
Pages462-470
Number of pages9
Volume2018-June
ISBN (Electronic)9781538661000
DOIs
StatePublished - Dec 13 2018
Event31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 - Salt Lake City, United States
Duration: Jun 18 2018Jun 22 2018

Other

Other31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018
CountryUnited States
CitySalt Lake City
Period6/18/186/22/18

Fingerprint

Accelerometers
Pipelines
Functional analysis
Human computer interaction
Learning systems
Experiments
Health
Sampling

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Choi, H., Wang, Q., Toledo, M., Turaga, P., Buman, M., & Srivastava, A. (2018). Temporal alignment improves feature quality: An experiment on activity recognition with accelerometer data. In Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 (Vol. 2018-June, pp. 462-470). [8575537] IEEE Computer Society. https://doi.org/10.1109/CVPRW.2018.00075

Temporal alignment improves feature quality : An experiment on activity recognition with accelerometer data. / Choi, Hongjun; Wang, Qiao; Toledo, Meynard; Turaga, Pavan; Buman, Matthew; Srivastava, Anuj.

Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018. Vol. 2018-June IEEE Computer Society, 2018. p. 462-470 8575537.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Choi, H, Wang, Q, Toledo, M, Turaga, P, Buman, M & Srivastava, A 2018, Temporal alignment improves feature quality: An experiment on activity recognition with accelerometer data. in Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018. vol. 2018-June, 8575537, IEEE Computer Society, pp. 462-470, 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018, Salt Lake City, United States, 6/18/18. https://doi.org/10.1109/CVPRW.2018.00075
Choi H, Wang Q, Toledo M, Turaga P, Buman M, Srivastava A. Temporal alignment improves feature quality: An experiment on activity recognition with accelerometer data. In Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018. Vol. 2018-June. IEEE Computer Society. 2018. p. 462-470. 8575537 https://doi.org/10.1109/CVPRW.2018.00075
Choi, Hongjun ; Wang, Qiao ; Toledo, Meynard ; Turaga, Pavan ; Buman, Matthew ; Srivastava, Anuj. / Temporal alignment improves feature quality : An experiment on activity recognition with accelerometer data. Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018. Vol. 2018-June IEEE Computer Society, 2018. pp. 462-470
@inproceedings{d3ae63fbf5474740ab87ff7e82f5673d,
title = "Temporal alignment improves feature quality: An experiment on activity recognition with accelerometer data",
abstract = "Activity recognition has been receiving significant attention from a variety of research areas such as human performance enhancement, health promotion, and human computer interaction. However, recognizing activities from accelerometer data still remains a challenging problem due to sensitivity to sampling rates, misalignment of data, and increased variability in activities among clinically relevant populations. In order to solve these issues, we adopt methods from functional analysis, which consider non-elastic rate variations in movement. The overall framework factors out temporal variability within activity classes, before leveraging robust machine learning pipelines for a given end-use. The proposed approach has been evaluated on 7 classes of everyday activities with 50 subjects. The results indicate that proposed approach achieves improved performance with the improvements observed in separating similar classes that differ in temporal rates, and also demonstrate higher robustness to change in window lengths. These results suggest that temporal alignment should be considered a core part of activity recognition pipelines.",
author = "Hongjun Choi and Qiao Wang and Meynard Toledo and Pavan Turaga and Matthew Buman and Anuj Srivastava",
year = "2018",
month = "12",
day = "13",
doi = "10.1109/CVPRW.2018.00075",
language = "English (US)",
volume = "2018-June",
pages = "462--470",
booktitle = "Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Temporal alignment improves feature quality

T2 - An experiment on activity recognition with accelerometer data

AU - Choi, Hongjun

AU - Wang, Qiao

AU - Toledo, Meynard

AU - Turaga, Pavan

AU - Buman, Matthew

AU - Srivastava, Anuj

PY - 2018/12/13

Y1 - 2018/12/13

N2 - Activity recognition has been receiving significant attention from a variety of research areas such as human performance enhancement, health promotion, and human computer interaction. However, recognizing activities from accelerometer data still remains a challenging problem due to sensitivity to sampling rates, misalignment of data, and increased variability in activities among clinically relevant populations. In order to solve these issues, we adopt methods from functional analysis, which consider non-elastic rate variations in movement. The overall framework factors out temporal variability within activity classes, before leveraging robust machine learning pipelines for a given end-use. The proposed approach has been evaluated on 7 classes of everyday activities with 50 subjects. The results indicate that proposed approach achieves improved performance with the improvements observed in separating similar classes that differ in temporal rates, and also demonstrate higher robustness to change in window lengths. These results suggest that temporal alignment should be considered a core part of activity recognition pipelines.

AB - Activity recognition has been receiving significant attention from a variety of research areas such as human performance enhancement, health promotion, and human computer interaction. However, recognizing activities from accelerometer data still remains a challenging problem due to sensitivity to sampling rates, misalignment of data, and increased variability in activities among clinically relevant populations. In order to solve these issues, we adopt methods from functional analysis, which consider non-elastic rate variations in movement. The overall framework factors out temporal variability within activity classes, before leveraging robust machine learning pipelines for a given end-use. The proposed approach has been evaluated on 7 classes of everyday activities with 50 subjects. The results indicate that proposed approach achieves improved performance with the improvements observed in separating similar classes that differ in temporal rates, and also demonstrate higher robustness to change in window lengths. These results suggest that temporal alignment should be considered a core part of activity recognition pipelines.

UR - http://www.scopus.com/inward/record.url?scp=85060897066&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060897066&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2018.00075

DO - 10.1109/CVPRW.2018.00075

M3 - Conference contribution

AN - SCOPUS:85060897066

VL - 2018-June

SP - 462

EP - 470

BT - Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018

PB - IEEE Computer Society

ER -