Loading [MathJax]/extensions/tex2jax.js

Unsupervised Features for Facial Expression Intensity Estimation over Time

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Autorschaft

Details

OriginalspracheEnglisch
Titel des SammelwerksProceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops
UntertitelCVPRW 2018
Herausgeber (Verlag)IEEE Computer Society
Seiten1199-1207
Seitenumfang9
ISBN (elektronisch)9781538661000
PublikationsstatusVeröffentlicht - 13 Dez. 2018
Veranstaltung20018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 - Salt Lake City, USA / Vereinigte Staaten
Dauer: 18 Juni 201823 Juni 2018

Publikationsreihe

NameIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Band2018-June
ISSN (Print)2160-7508
ISSN (elektronisch)2160-7516

Abstract

The diversity of facial shapes and motions among persons is one of the greatest challenges for automatic analysis of facial expressions. In this paper, we propose a feature describing expression intensity over time, while being invariant to person and the type of performed expression. Our feature is a weighted combination of the dynamics of multiple points adapted to the overall expression trajectory. We evaluate our method on several tasks all related to temporal analysis of facial expression. The proposed feature is compared to a state-of-the-art method for expression intensity estimation, which it outperforms. We use our proposed feature to temporally align multiple sequences of recorded 3D facial expressions. Furthermore, we show how our feature can be used to reveal person-specific differences in performances of facial expressions. Additionally, we apply our feature to identify the local changes in face video sequences based on action unit labels. For all the experiments our feature proves to be robust against noise and outliers, making it applicable to a variety of applications for analysis of facial movements.

ASJC Scopus Sachgebiete

Zitieren

Unsupervised Features for Facial Expression Intensity Estimation over Time. / Awiszus, Maren; Grabhof, Stella; Kuhnke, Felix et al.
Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2018. IEEE Computer Society, 2018. S. 1199-1207 8575310 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Band 2018-June).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Awiszus, M, Grabhof, S, Kuhnke, F & Ostermann, J 2018, Unsupervised Features for Facial Expression Intensity Estimation over Time. in Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2018., 8575310, IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Bd. 2018-June, IEEE Computer Society, S. 1199-1207, 20018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, Utah, USA / Vereinigte Staaten, 18 Juni 2018. https://doi.org/arXiv:1805.00780v2
Awiszus, M., Grabhof, S., Kuhnke, F., & Ostermann, J. (2018). Unsupervised Features for Facial Expression Intensity Estimation over Time. In Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2018 (S. 1199-1207). Artikel 8575310 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Band 2018-June). IEEE Computer Society. https://doi.org/arXiv:1805.00780v2
Awiszus M, Grabhof S, Kuhnke F, Ostermann J. Unsupervised Features for Facial Expression Intensity Estimation over Time. in Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2018. IEEE Computer Society. 2018. S. 1199-1207. 8575310. (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). doi: arXiv:1805.00780v2
Awiszus, Maren ; Grabhof, Stella ; Kuhnke, Felix et al. / Unsupervised Features for Facial Expression Intensity Estimation over Time. Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2018. IEEE Computer Society, 2018. S. 1199-1207 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).
Download
@inproceedings{9e1b4a2c8e7442ec98540bcce4ebd4cf,
title = "Unsupervised Features for Facial Expression Intensity Estimation over Time",
abstract = "The diversity of facial shapes and motions among persons is one of the greatest challenges for automatic analysis of facial expressions. In this paper, we propose a feature describing expression intensity over time, while being invariant to person and the type of performed expression. Our feature is a weighted combination of the dynamics of multiple points adapted to the overall expression trajectory. We evaluate our method on several tasks all related to temporal analysis of facial expression. The proposed feature is compared to a state-of-the-art method for expression intensity estimation, which it outperforms. We use our proposed feature to temporally align multiple sequences of recorded 3D facial expressions. Furthermore, we show how our feature can be used to reveal person-specific differences in performances of facial expressions. Additionally, we apply our feature to identify the local changes in face video sequences based on action unit labels. For all the experiments our feature proves to be robust against noise and outliers, making it applicable to a variety of applications for analysis of facial movements.",
author = "Maren Awiszus and Stella Grabhof and Felix Kuhnke and Jorn Ostermann",
year = "2018",
month = dec,
day = "13",
doi = "arXiv:1805.00780v2",
language = "English",
series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",
publisher = "IEEE Computer Society",
pages = "1199--1207",
booktitle = "Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops",
address = "United States",
note = "20018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018, CVPR 2018 ; Conference date: 18-06-2018 Through 23-06-2018",

}

Download

TY - GEN

T1 - Unsupervised Features for Facial Expression Intensity Estimation over Time

AU - Awiszus, Maren

AU - Grabhof, Stella

AU - Kuhnke, Felix

AU - Ostermann, Jorn

PY - 2018/12/13

Y1 - 2018/12/13

N2 - The diversity of facial shapes and motions among persons is one of the greatest challenges for automatic analysis of facial expressions. In this paper, we propose a feature describing expression intensity over time, while being invariant to person and the type of performed expression. Our feature is a weighted combination of the dynamics of multiple points adapted to the overall expression trajectory. We evaluate our method on several tasks all related to temporal analysis of facial expression. The proposed feature is compared to a state-of-the-art method for expression intensity estimation, which it outperforms. We use our proposed feature to temporally align multiple sequences of recorded 3D facial expressions. Furthermore, we show how our feature can be used to reveal person-specific differences in performances of facial expressions. Additionally, we apply our feature to identify the local changes in face video sequences based on action unit labels. For all the experiments our feature proves to be robust against noise and outliers, making it applicable to a variety of applications for analysis of facial movements.

AB - The diversity of facial shapes and motions among persons is one of the greatest challenges for automatic analysis of facial expressions. In this paper, we propose a feature describing expression intensity over time, while being invariant to person and the type of performed expression. Our feature is a weighted combination of the dynamics of multiple points adapted to the overall expression trajectory. We evaluate our method on several tasks all related to temporal analysis of facial expression. The proposed feature is compared to a state-of-the-art method for expression intensity estimation, which it outperforms. We use our proposed feature to temporally align multiple sequences of recorded 3D facial expressions. Furthermore, we show how our feature can be used to reveal person-specific differences in performances of facial expressions. Additionally, we apply our feature to identify the local changes in face video sequences based on action unit labels. For all the experiments our feature proves to be robust against noise and outliers, making it applicable to a variety of applications for analysis of facial movements.

UR - http://www.scopus.com/inward/record.url?scp=85060868946&partnerID=8YFLogxK

U2 - arXiv:1805.00780v2

DO - arXiv:1805.00780v2

M3 - Conference contribution

AN - SCOPUS:85060868946

T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

SP - 1199

EP - 1207

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

PB - IEEE Computer Society

T2 - 20018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018

Y2 - 18 June 2018 through 23 June 2018

ER -

Von denselben Autoren