Data Trimming Methods to Improve Gesture Classification

Hye Sung Roh; Dae Eun Kim

doi:10.23919/ICEMS52562.2021.9634650

Data Trimming Methods to Improve Gesture Classification

Hye Sung Roh, Dae Eun Kim

Department of Electrical and Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper introduces a data processing method to enhance the performance of a gesture classification model. When tested on the UTD-MHAD dataset, the HMM model initially rendered a poor performance due to seemingly resembling gestures. To tackle this problem, data has been altered via normalization and selection of significant joints that determine the gesture. Refining data prior to classifying generates a better performance in both HMM and LSTM models, highlighting the significance of data processing across different types of classification models.

Original language	English
Title of host publication	ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2449-2452
Number of pages	4
ISBN (Electronic)	9788986510218
DOIs	https://doi.org/10.23919/ICEMS52562.2021.9634650
Publication status	Published - 2021
Event	24th International Conference on Electrical Machines and Systems, ICEMS 2021 - Gyeongju, Korea, Republic of Duration: 2021 Oct 31 → 2021 Nov 3

Publication series

Name	ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems

Conference

Conference	24th International Conference on Electrical Machines and Systems, ICEMS 2021
Country/Territory	Korea, Republic of
City	Gyeongju
Period	21/10/31 → 21/11/3

Bibliographical note

Publisher Copyright:
© 2021 KIEE & EMECS.

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering
Mechanical Engineering
Safety, Risk, Reliability and Quality

Access to Document

10.23919/ICEMS52562.2021.9634650

Cite this

Roh, H. S., & Kim, D. E. (2021). Data Trimming Methods to Improve Gesture Classification. In ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems (pp. 2449-2452). (ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/ICEMS52562.2021.9634650

@inproceedings{e0ebf9ac18c544f7b078c47038c88024,

title = "Data Trimming Methods to Improve Gesture Classification",

abstract = "This paper introduces a data processing method to enhance the performance of a gesture classification model. When tested on the UTD-MHAD dataset, the HMM model initially rendered a poor performance due to seemingly resembling gestures. To tackle this problem, data has been altered via normalization and selection of significant joints that determine the gesture. Refining data prior to classifying generates a better performance in both HMM and LSTM models, highlighting the significance of data processing across different types of classification models.",

keywords = "Data Preprocessing, Gesture Recognition, Hidden Markov Model (HMM), Human-Computer Interaction",

author = "Roh, {Hye Sung} and Kim, {Dae Eun}",

note = "Publisher Copyright: {\textcopyright} 2021 KIEE & EMECS.; 24th International Conference on Electrical Machines and Systems, ICEMS 2021 ; Conference date: 31-10-2021 Through 03-11-2021",

year = "2021",

doi = "10.23919/ICEMS52562.2021.9634650",

language = "English",

series = "ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "2449--2452",

booktitle = "ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems",

address = "United States",

}

Roh, HS & Kim, DE 2021, Data Trimming Methods to Improve Gesture Classification. in ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems. ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems, Institute of Electrical and Electronics Engineers Inc., pp. 2449-2452, 24th International Conference on Electrical Machines and Systems, ICEMS 2021, Gyeongju, Korea, Republic of, 21/10/31. https://doi.org/10.23919/ICEMS52562.2021.9634650

Data Trimming Methods to Improve Gesture Classification. / Roh, Hye Sung; Kim, Dae Eun.
ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems. Institute of Electrical and Electronics Engineers Inc., 2021. p. 2449-2452 (ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Data Trimming Methods to Improve Gesture Classification

AU - Roh, Hye Sung

AU - Kim, Dae Eun

PY - 2021

Y1 - 2021

N2 - This paper introduces a data processing method to enhance the performance of a gesture classification model. When tested on the UTD-MHAD dataset, the HMM model initially rendered a poor performance due to seemingly resembling gestures. To tackle this problem, data has been altered via normalization and selection of significant joints that determine the gesture. Refining data prior to classifying generates a better performance in both HMM and LSTM models, highlighting the significance of data processing across different types of classification models.

AB - This paper introduces a data processing method to enhance the performance of a gesture classification model. When tested on the UTD-MHAD dataset, the HMM model initially rendered a poor performance due to seemingly resembling gestures. To tackle this problem, data has been altered via normalization and selection of significant joints that determine the gesture. Refining data prior to classifying generates a better performance in both HMM and LSTM models, highlighting the significance of data processing across different types of classification models.

KW - Data Preprocessing

KW - Gesture Recognition

KW - Hidden Markov Model (HMM)

KW - Human-Computer Interaction

UR - http://www.scopus.com/inward/record.url?scp=85123911087&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85123911087&partnerID=8YFLogxK

U2 - 10.23919/ICEMS52562.2021.9634650

DO - 10.23919/ICEMS52562.2021.9634650

M3 - Conference contribution

AN - SCOPUS:85123911087

T3 - ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems

SP - 2449

EP - 2452

BT - ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 24th International Conference on Electrical Machines and Systems, ICEMS 2021

Y2 - 31 October 2021 through 3 November 2021

ER -

Data Trimming Methods to Improve Gesture Classification

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this