A temporally piece-wise fisher vector approach for depression analysis

Abhinav DHALL, Roland GOECKE

Research output: A Conference proceeding or a Chapter in BookConference contribution

14 Citations (Scopus)

Abstract

Depression and other mood disorders are common, disabling disorders with a profound impact on individuals and families. Inspite of its high prevalence, it is easily missed during the early stages. Automatic depression analysis has become a very active field of research in the affective computing community in the past few years. This paper presents a framework for
depression analysis based on unimodal visual cues. Temporally piece-wise Fisher Vectors (FV) are computed on temporal segments. As a low-level feature, block-wise Local Binary Pattern-Three Orthogonal Planes descriptors are computed. Statistical aggregation techniques are analysed and compared for creating a discriminative representative for a video sample. The paper explores the strength of FV in representing temporal segments in a spontaneous clinical data. This creates a meaningful representation of the facial dynamics in a temporal segment. The experiments are conducted on the Audio Video Emotion Challenge (AVEC) 2014 German speaking depression database. The superior results of the proposed framework show the effectiveness of the technique as compared to the current state-of-art
Original languageEnglish
Title of host publication2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015
EditorsRoddy Lowie, Qiang Ji, Jianhua Tao
Place of PublicationChina
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages255-259
Number of pages5
Volume1
ISBN (Print)9781479999538, 9781479999521
DOIs
Publication statusPublished - 2015
Event6th International conference on affective computing and intelligent interfaces - Xi-an, Xi-an, China
Duration: 21 Sep 201524 Sep 2015

Publication series

Name2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015

Conference

Conference6th International conference on affective computing and intelligent interfaces
Abbreviated titleACII 2015
CountryChina
CityXi-an
Period21/09/1524/09/15

Fingerprint

Agglomeration
Experiments

Cite this

DHALL, A., & GOECKE, R. (2015). A temporally piece-wise fisher vector approach for depression analysis. In R. Lowie, Q. Ji, & J. Tao (Eds.), 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015 (Vol. 1, pp. 255-259). [7344580] (2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015). China: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ACII.2015.7344580
DHALL, Abhinav ; GOECKE, Roland. / A temporally piece-wise fisher vector approach for depression analysis. 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015. editor / Roddy Lowie ; Qiang Ji ; Jianhua Tao. Vol. 1 China : IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 255-259 (2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015).
@inproceedings{0efb135a7fa445109d6952ba6c0db686,
title = "A temporally piece-wise fisher vector approach for depression analysis",
abstract = "Depression and other mood disorders are common, disabling disorders with a profound impact on individuals and families. Inspite of its high prevalence, it is easily missed during the early stages. Automatic depression analysis has become a very active field of research in the affective computing community in the past few years. This paper presents a framework fordepression analysis based on unimodal visual cues. Temporally piece-wise Fisher Vectors (FV) are computed on temporal segments. As a low-level feature, block-wise Local Binary Pattern-Three Orthogonal Planes descriptors are computed. Statistical aggregation techniques are analysed and compared for creating a discriminative representative for a video sample. The paper explores the strength of FV in representing temporal segments in a spontaneous clinical data. This creates a meaningful representation of the facial dynamics in a temporal segment. The experiments are conducted on the Audio Video Emotion Challenge (AVEC) 2014 German speaking depression database. The superior results of the proposed framework show the effectiveness of the technique as compared to the current state-of-art",
keywords = "Depression Analysis, Fisher Vector",
author = "Abhinav DHALL and Roland GOECKE",
year = "2015",
doi = "10.1109/ACII.2015.7344580",
language = "English",
isbn = "9781479999538",
volume = "1",
series = "2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "255--259",
editor = "Roddy Lowie and Qiang Ji and Jianhua Tao",
booktitle = "2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015",
address = "United States",

}

DHALL, A & GOECKE, R 2015, A temporally piece-wise fisher vector approach for depression analysis. in R Lowie, Q Ji & J Tao (eds), 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015. vol. 1, 7344580, 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015, IEEE, Institute of Electrical and Electronics Engineers, China, pp. 255-259, 6th International conference on affective computing and intelligent interfaces, Xi-an, China, 21/09/15. https://doi.org/10.1109/ACII.2015.7344580

A temporally piece-wise fisher vector approach for depression analysis. / DHALL, Abhinav; GOECKE, Roland.

2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015. ed. / Roddy Lowie; Qiang Ji; Jianhua Tao. Vol. 1 China : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 255-259 7344580 (2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - A temporally piece-wise fisher vector approach for depression analysis

AU - DHALL, Abhinav

AU - GOECKE, Roland

PY - 2015

Y1 - 2015

N2 - Depression and other mood disorders are common, disabling disorders with a profound impact on individuals and families. Inspite of its high prevalence, it is easily missed during the early stages. Automatic depression analysis has become a very active field of research in the affective computing community in the past few years. This paper presents a framework fordepression analysis based on unimodal visual cues. Temporally piece-wise Fisher Vectors (FV) are computed on temporal segments. As a low-level feature, block-wise Local Binary Pattern-Three Orthogonal Planes descriptors are computed. Statistical aggregation techniques are analysed and compared for creating a discriminative representative for a video sample. The paper explores the strength of FV in representing temporal segments in a spontaneous clinical data. This creates a meaningful representation of the facial dynamics in a temporal segment. The experiments are conducted on the Audio Video Emotion Challenge (AVEC) 2014 German speaking depression database. The superior results of the proposed framework show the effectiveness of the technique as compared to the current state-of-art

AB - Depression and other mood disorders are common, disabling disorders with a profound impact on individuals and families. Inspite of its high prevalence, it is easily missed during the early stages. Automatic depression analysis has become a very active field of research in the affective computing community in the past few years. This paper presents a framework fordepression analysis based on unimodal visual cues. Temporally piece-wise Fisher Vectors (FV) are computed on temporal segments. As a low-level feature, block-wise Local Binary Pattern-Three Orthogonal Planes descriptors are computed. Statistical aggregation techniques are analysed and compared for creating a discriminative representative for a video sample. The paper explores the strength of FV in representing temporal segments in a spontaneous clinical data. This creates a meaningful representation of the facial dynamics in a temporal segment. The experiments are conducted on the Audio Video Emotion Challenge (AVEC) 2014 German speaking depression database. The superior results of the proposed framework show the effectiveness of the technique as compared to the current state-of-art

KW - Depression Analysis

KW - Fisher Vector

UR - http://www.scopus.com/inward/record.url?scp=84964053689&partnerID=8YFLogxK

U2 - 10.1109/ACII.2015.7344580

DO - 10.1109/ACII.2015.7344580

M3 - Conference contribution

SN - 9781479999538

SN - 9781479999521

VL - 1

T3 - 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015

SP - 255

EP - 259

BT - 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015

A2 - Lowie, Roddy

A2 - Ji, Qiang

A2 - Tao, Jianhua

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - China

ER -

DHALL A, GOECKE R. A temporally piece-wise fisher vector approach for depression analysis. In Lowie R, Ji Q, Tao J, editors, 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015. Vol. 1. China: IEEE, Institute of Electrical and Electronics Engineers. 2015. p. 255-259. 7344580. (2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015). https://doi.org/10.1109/ACII.2015.7344580