Move, and I will tell you who you are: Detecting deceptive roles in low-quality data

Nimrod Raiman; Hayley Hung; Gwenn Englebienne

doi:10.1145/2070481.2070515

Move, and I will tell you who you are: Detecting deceptive roles in low-quality data

Nimrod Raiman^*
, Hayley Hung
, Gwenn Englebienne

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

13 Citations (Scopus)

Abstract

Motion, like speech, provides information about one's emotional state. This work introduces an automated non-verbal audio-visual approach for detecting deceptive roles in multi-party conversations using low resolution video. We show how using simple features extracted from motion and speech improves over speech-only for the detection of deceptive roles. Our results show that deceptive players were recognised with significantly higher precision when video features were used. We improve the classification performance with 22.6% compared to our baseline.

Original language	English
Title of host publication	ICMI'11
Subtitle of host publication	Proceedings of the 2011 ACM International Conference on Multimodal Interaction
Editors	Hervé Bourlard
Publisher	ACM Publishing
Pages	201-204
Number of pages	4
ISBN (Print)	978-1-4503-0641-6
DOIs	https://doi.org/10.1145/2070481.2070515
Publication status	Published - 2011
Externally published	Yes
Event	13th International Conference on Multimodal Interfaces, ICMI 2011 - Alicante, Spain Duration: 14 Nov 2011 → 18 Nov 2011 Conference number: 13

Conference

Conference	13th International Conference on Multimodal Interfaces, ICMI 2011
Abbreviated title	ICMI
Country/Territory	Spain
City	Alicante
Period	14/11/11 → 18/11/11

Keywords

Deception detection
Human behavior
Multi-party conversation
n/a OA procedure

Access to Document

10.1145/2070481.2070515

Cite this

@inproceedings{6a5a58ac4f8040dda8eab348b2cf9c6c,

title = "Move, and I will tell you who you are: Detecting deceptive roles in low-quality data",

abstract = "Motion, like speech, provides information about one's emotional state. This work introduces an automated non-verbal audio-visual approach for detecting deceptive roles in multi-party conversations using low resolution video. We show how using simple features extracted from motion and speech improves over speech-only for the detection of deceptive roles. Our results show that deceptive players were recognised with significantly higher precision when video features were used. We improve the classification performance with 22.6\% compared to our baseline.",

keywords = "Deception detection, Human behavior, Multi-party conversation, n/a OA procedure",

author = "Nimrod Raiman and Hayley Hung and Gwenn Englebienne",

year = "2011",

doi = "10.1145/2070481.2070515",

language = "English",

isbn = "978-1-4503-0641-6",

pages = "201--204",

editor = "Herv{\'e} Bourlard",

booktitle = "ICMI'11",

publisher = "ACM Publishing",

note = "13th International Conference on Multimodal Interfaces, ICMI 2011, ICMI ; Conference date: 14-11-2011 Through 18-11-2011",

}

Raiman, N, Hung, H & Englebienne, G 2011, Move, and I will tell you who you are: Detecting deceptive roles in low-quality data. in H Bourlard (ed.), ICMI'11: Proceedings of the 2011 ACM International Conference on Multimodal Interaction. ACM Publishing, pp. 201-204, 13th International Conference on Multimodal Interfaces, ICMI 2011, Alicante, Spain, 14/11/11. https://doi.org/10.1145/2070481.2070515

Move, and I will tell you who you are: Detecting deceptive roles in low-quality data. / Raiman, Nimrod; Hung, Hayley; Englebienne, Gwenn.
ICMI'11: Proceedings of the 2011 ACM International Conference on Multimodal Interaction. ed. / Hervé Bourlard. ACM Publishing, 2011. p. 201-204.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Move, and I will tell you who you are

T2 - 13th International Conference on Multimodal Interfaces, ICMI 2011

AU - Raiman, Nimrod

AU - Hung, Hayley

AU - Englebienne, Gwenn

N1 - Conference code: 13

PY - 2011

Y1 - 2011

N2 - Motion, like speech, provides information about one's emotional state. This work introduces an automated non-verbal audio-visual approach for detecting deceptive roles in multi-party conversations using low resolution video. We show how using simple features extracted from motion and speech improves over speech-only for the detection of deceptive roles. Our results show that deceptive players were recognised with significantly higher precision when video features were used. We improve the classification performance with 22.6% compared to our baseline.

AB - Motion, like speech, provides information about one's emotional state. This work introduces an automated non-verbal audio-visual approach for detecting deceptive roles in multi-party conversations using low resolution video. We show how using simple features extracted from motion and speech improves over speech-only for the detection of deceptive roles. Our results show that deceptive players were recognised with significantly higher precision when video features were used. We improve the classification performance with 22.6% compared to our baseline.

KW - Deception detection

KW - Human behavior

KW - Multi-party conversation

KW - n/a OA procedure

UR - https://www.scopus.com/pages/publications/83455176746

U2 - 10.1145/2070481.2070515

DO - 10.1145/2070481.2070515

M3 - Conference contribution

AN - SCOPUS:83455176746

SN - 978-1-4503-0641-6

SP - 201

EP - 204

BT - ICMI'11

A2 - Bourlard, Hervé

PB - ACM Publishing

Y2 - 14 November 2011 through 18 November 2011

ER -

Move, and I will tell you who you are: Detecting deceptive roles in low-quality data

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this