A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022

Shurong Chai; Jiaqing Liu; Rahul Kumar Jain; Yinhao Li; Tomoko Tateyama; Yen Wei Chen

doi:10.1109/PerComWorkshops56833.2023.10150404

A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022

Shurong Chai, Jiaqing Liu, Rahul Kumar Jain, Yinhao Li, Tomoko Tateyama, Yen Wei Chen

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

2 Citations (Scopus)

Abstract

We report the proposed method of Team Ritsumei for the OpenPack challenge 2022. In this work, we proposed to use a motion-aware and temporal-enhanced Spatial-Temporal Graph Convolutional Networks for the representation of the keypoint modality features. We also leverage the Accelerometer and Gyroscope modality as auxiliary modalities to improve the performance. Our final result is based on the fusion of four modalities. We report the 92.32% F1 score on the submission set, which won the 3rd place in the OpenPack challenge 2022.

Original language	English
Title of host publication	Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	267-269
Number of pages	3
ISBN (Electronic)	9781665453813
DOIs	https://doi.org/10.1109/PerComWorkshops56833.2023.10150404
Publication status	Published - 2023
Externally published	Yes
Event	21st IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023 - Atlanta, United States Duration: 13-03-2023 → 17-03-2023

Publication series

Name	2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023

Conference

Conference	21st IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023
Country/Territory	United States
City	Atlanta
Period	13-03-23 → 17-03-23

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Computer Science Applications
Information Systems
Information Systems and Management
Health Informatics
Psychology (miscellaneous)

Access to Document

10.1109/PerComWorkshops56833.2023.10150404

Cite this

Chai, S., Liu, J., Jain, R. K., Li, Y., Tateyama, T., & Chen, Y. W. (2023). A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022. In Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023 (pp. 267-269). (2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PerComWorkshops56833.2023.10150404

Chai, Shurong ; Liu, Jiaqing ; Jain, Rahul Kumar et al. / A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022. Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 267-269 (2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023).

@inproceedings{d6d01dc05a1c4136837e97e888041ab7,

title = "A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022",

abstract = "We report the proposed method of Team Ritsumei for the OpenPack challenge 2022. In this work, we proposed to use a motion-aware and temporal-enhanced Spatial-Temporal Graph Convolutional Networks for the representation of the keypoint modality features. We also leverage the Accelerometer and Gyroscope modality as auxiliary modalities to improve the performance. Our final result is based on the fusion of four modalities. We report the 92.32\% F1 score on the submission set, which won the 3rd place in the OpenPack challenge 2022.",

keywords = "Action segmentation, OpenPack challenge, Spatial-Temporal Graph Convolutional Networks",

author = "Shurong Chai and Jiaqing Liu and Jain, \{Rahul Kumar\} and Yinhao Li and Tomoko Tateyama and Chen, \{Yen Wei\}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 21st IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023 ; Conference date: 13-03-2023 Through 17-03-2023",

year = "2023",

doi = "10.1109/PerComWorkshops56833.2023.10150404",

language = "English",

series = "2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023",

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

pages = "267--269",

booktitle = "Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023",

address = "United States",

}

Chai, S, Liu, J, Jain, RK, Li, Y, Tateyama, T & Chen, YW 2023, A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022. in Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023. 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023, Institute of Electrical and Electronics Engineers Inc., pp. 267-269, 21st IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023, Atlanta, United States, 13-03-23. https://doi.org/10.1109/PerComWorkshops56833.2023.10150404

A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022. / Chai, Shurong; Liu, Jiaqing; Jain, Rahul Kumar et al.
Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 267-269 (2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023).

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

TY - GEN

T1 - A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022

AU - Chai, Shurong

AU - Liu, Jiaqing

AU - Jain, Rahul Kumar

AU - Li, Yinhao

AU - Tateyama, Tomoko

AU - Chen, Yen Wei

PY - 2023

Y1 - 2023

N2 - We report the proposed method of Team Ritsumei for the OpenPack challenge 2022. In this work, we proposed to use a motion-aware and temporal-enhanced Spatial-Temporal Graph Convolutional Networks for the representation of the keypoint modality features. We also leverage the Accelerometer and Gyroscope modality as auxiliary modalities to improve the performance. Our final result is based on the fusion of four modalities. We report the 92.32% F1 score on the submission set, which won the 3rd place in the OpenPack challenge 2022.

AB - We report the proposed method of Team Ritsumei for the OpenPack challenge 2022. In this work, we proposed to use a motion-aware and temporal-enhanced Spatial-Temporal Graph Convolutional Networks for the representation of the keypoint modality features. We also leverage the Accelerometer and Gyroscope modality as auxiliary modalities to improve the performance. Our final result is based on the fusion of four modalities. We report the 92.32% F1 score on the submission set, which won the 3rd place in the OpenPack challenge 2022.

KW - Action segmentation

KW - OpenPack challenge

KW - Spatial-Temporal Graph Convolutional Networks

UR - https://www.scopus.com/pages/publications/85164147888

UR - https://www.scopus.com/inward/citedby.url?scp=85164147888&partnerID=8YFLogxK

U2 - 10.1109/PerComWorkshops56833.2023.10150404

DO - 10.1109/PerComWorkshops56833.2023.10150404

M3 - Conference contribution

AN - SCOPUS:85164147888

T3 - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023

SP - 267

EP - 269

BT - Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 21st IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023

Y2 - 13 March 2023 through 17 March 2023

ER -

Chai S, Liu J, Jain RK, Li Y, Tateyama T, Chen YW. A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022. In Proceedings - 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 267-269. (2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023). doi: 10.1109/PerComWorkshops56833.2023.10150404

A Spatial-Temporal Graph Convolutional Networks-based Approach for the OpenPack Challenge 2022

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this