ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results

Cristina Palmero; German Barquero; Julio C.S. Jacques Junior; Albert Clapés; Johnny Núñez; David Curto; Sorina Smeureanu; Javier Selva; Zejian Zhang; David Saeteros; David Gallardo-Pujol; Georgina Guilera; David Leiva; Feng Han; Xiaoxue Feng; Jennifer He; Wei Wei Tu; Thomas B. Moeslund; Isabelle Guyon; Sergio Escalera

ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results

Cristina Palmero, German Barquero, Julio C.S. Jacques Junior, Albert Clapés, Johnny Núñez, David Curto, Sorina Smeureanu, Javier Selva, Zejian Zhang, David Saeteros, David Gallardo-Pujol, Georgina Guilera, David Leiva, Feng Han, Xiaoxue Feng, Jennifer He, Wei Wei Tu, Thomas B. Moeslund, Isabelle Guyon, Sergio Escalera

Research output: Contribution to journal › Conference article in Journal › Research › peer-review

12 Citations (Scopus)

Abstract

This paper summarizes the 2021 ChaLearn Looking at People Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions (DYAD), which featured two tracks, self-reported personality recognition and behavior forecasting, both on the UDIVA v0.5 dataset. We review important aspects of this multimodal and multiview dataset consisting of 145 interaction sessions where 134 participants converse, collaborate, and compete in a series of dyadic tasks. We also detail the transcripts and body landmark annotations for UDIVA v0.5 that are newly introduced for this occasion. We briefly comment on organizational aspects of the challenge before describing each track and presenting the proposed baselines. The results obtained by the participants are extensively analyzed to bring interesting insights about the tracks tasks and the nature of the dataset. We wrap up with a discussion on challenge outcomes, and pose several questions that we expect will motivate further scientific research to better understand social cues in human-human and human-machine interaction scenarios and help build future AI applications for good.

Original language	English
Book series	Proceedings of Machine Learning Research
Volume	173
Pages (from-to)	4-52
Number of pages	49
ISSN	2640-3498
Publication status	Published - 2021
Event	ChaLearn LAP Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions Workshop, DYAD 2021, held in conjunction with the International Conference on Computer Vision, ICCV 2021 - Virtual, Online Duration: 16 Oct 2021 → …

Conference

Conference	ChaLearn LAP Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions Workshop, DYAD 2021, held in conjunction with the International Conference on Computer Vision, ICCV 2021
City	Virtual, Online
Period	16/10/2021 → …

Bibliographical note

Funding Information:
We acknowledge ChaLearn and 4Paradigm for their support on annotating the dataset and for sponsoring the DYAD workshop and challenge. We acknowledge Meta Reality Labs and 4Paradigm for sponsoring the DYAD challenge prizes. We thank all challenge participants and researchers that requested dataset access for their interest in the challenge. This research was supported by Spanish project PID2019-105093GB-I00, ICREA under the ICREA Academia program, and ANR Chair of Artificial Intelligence HUMANIA ANR-19-CHIA-0022.

Publisher Copyright:
© 2022 C. Palmero et al.

Keywords

AI competitions
Behavior forecasting
Human interaction
Multimodal approaches
Personality recognition

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Palmero, C., Barquero, G., Jacques Junior, J. C. S., Clapés, A., Núñez, J., Curto, D., Smeureanu, S., Selva, J., Zhang, Z., Saeteros, D., Gallardo-Pujol, D., Guilera, G., Leiva, D., Han, F., Feng, X., He, J., Tu, W. W., Moeslund, T. B., Guyon, I., & Escalera, S. (2021). ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results. Proceedings of Machine Learning Research, 173, 4-52.

@inproceedings{e736c52b5d8745c09df4366345d9a448,

title = "ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results",

abstract = "This paper summarizes the 2021 ChaLearn Looking at People Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions (DYAD), which featured two tracks, self-reported personality recognition and behavior forecasting, both on the UDIVA v0.5 dataset. We review important aspects of this multimodal and multiview dataset consisting of 145 interaction sessions where 134 participants converse, collaborate, and compete in a series of dyadic tasks. We also detail the transcripts and body landmark annotations for UDIVA v0.5 that are newly introduced for this occasion. We briefly comment on organizational aspects of the challenge before describing each track and presenting the proposed baselines. The results obtained by the participants are extensively analyzed to bring interesting insights about the tracks tasks and the nature of the dataset. We wrap up with a discussion on challenge outcomes, and pose several questions that we expect will motivate further scientific research to better understand social cues in human-human and human-machine interaction scenarios and help build future AI applications for good.",

keywords = "AI competitions, Behavior forecasting, Human interaction, Multimodal approaches, Personality recognition",

author = "Cristina Palmero and German Barquero and {Jacques Junior}, {Julio C.S.} and Albert Clap{\'e}s and Johnny N{\'u}{\~n}ez and David Curto and Sorina Smeureanu and Javier Selva and Zejian Zhang and David Saeteros and David Gallardo-Pujol and Georgina Guilera and David Leiva and Feng Han and Xiaoxue Feng and Jennifer He and Tu, {Wei Wei} and Moeslund, {Thomas B.} and Isabelle Guyon and Sergio Escalera",

note = "Funding Information: We acknowledge ChaLearn and 4Paradigm for their support on annotating the dataset and for sponsoring the DYAD workshop and challenge. We acknowledge Meta Reality Labs and 4Paradigm for sponsoring the DYAD challenge prizes. We thank all challenge participants and researchers that requested dataset access for their interest in the challenge. This research was supported by Spanish project PID2019-105093GB-I00, ICREA under the ICREA Academia program, and ANR Chair of Artificial Intelligence HUMANIA ANR-19-CHIA-0022. Publisher Copyright: {\textcopyright} 2022 C. Palmero et al.; ChaLearn LAP Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions Workshop, DYAD 2021, held in conjunction with the International Conference on Computer Vision, ICCV 2021 ; Conference date: 16-10-2021",

year = "2021",

language = "English",

volume = "173",

pages = "4--52",

journal = "Proceedings of Machine Learning Research",

issn = "2640-3498",

}

Palmero, C, Barquero, G, Jacques Junior, JCS, Clapés, A, Núñez, J, Curto, D, Smeureanu, S, Selva, J, Zhang, Z, Saeteros, D, Gallardo-Pujol, D, Guilera, G, Leiva, D, Han, F, Feng, X, He, J, Tu, WW, Moeslund, TB, Guyon, I & Escalera, S 2021, 'ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results', Proceedings of Machine Learning Research, vol. 173, pp. 4-52.

ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results. / Palmero, Cristina; Barquero, German; Jacques Junior, Julio C.S. et al.
In: Proceedings of Machine Learning Research, Vol. 173, 2021, p. 4-52.

Research output: Contribution to journal › Conference article in Journal › Research › peer-review

TY - GEN

T1 - ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions

T2 - ChaLearn LAP Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions Workshop, DYAD 2021, held in conjunction with the International Conference on Computer Vision, ICCV 2021

AU - Palmero, Cristina

AU - Barquero, German

AU - Jacques Junior, Julio C.S.

AU - Clapés, Albert

AU - Núñez, Johnny

AU - Curto, David

AU - Smeureanu, Sorina

AU - Selva, Javier

AU - Zhang, Zejian

AU - Saeteros, David

AU - Gallardo-Pujol, David

AU - Guilera, Georgina

AU - Leiva, David

AU - Han, Feng

AU - Feng, Xiaoxue

AU - He, Jennifer

AU - Tu, Wei Wei

AU - Moeslund, Thomas B.

AU - Guyon, Isabelle

AU - Escalera, Sergio

N1 - Funding Information: We acknowledge ChaLearn and 4Paradigm for their support on annotating the dataset and for sponsoring the DYAD workshop and challenge. We acknowledge Meta Reality Labs and 4Paradigm for sponsoring the DYAD challenge prizes. We thank all challenge participants and researchers that requested dataset access for their interest in the challenge. This research was supported by Spanish project PID2019-105093GB-I00, ICREA under the ICREA Academia program, and ANR Chair of Artificial Intelligence HUMANIA ANR-19-CHIA-0022. Publisher Copyright: © 2022 C. Palmero et al.

PY - 2021

Y1 - 2021

N2 - This paper summarizes the 2021 ChaLearn Looking at People Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions (DYAD), which featured two tracks, self-reported personality recognition and behavior forecasting, both on the UDIVA v0.5 dataset. We review important aspects of this multimodal and multiview dataset consisting of 145 interaction sessions where 134 participants converse, collaborate, and compete in a series of dyadic tasks. We also detail the transcripts and body landmark annotations for UDIVA v0.5 that are newly introduced for this occasion. We briefly comment on organizational aspects of the challenge before describing each track and presenting the proposed baselines. The results obtained by the participants are extensively analyzed to bring interesting insights about the tracks tasks and the nature of the dataset. We wrap up with a discussion on challenge outcomes, and pose several questions that we expect will motivate further scientific research to better understand social cues in human-human and human-machine interaction scenarios and help build future AI applications for good.

AB - This paper summarizes the 2021 ChaLearn Looking at People Challenge on Understanding Social Behavior in Dyadic and Small Group Interactions (DYAD), which featured two tracks, self-reported personality recognition and behavior forecasting, both on the UDIVA v0.5 dataset. We review important aspects of this multimodal and multiview dataset consisting of 145 interaction sessions where 134 participants converse, collaborate, and compete in a series of dyadic tasks. We also detail the transcripts and body landmark annotations for UDIVA v0.5 that are newly introduced for this occasion. We briefly comment on organizational aspects of the challenge before describing each track and presenting the proposed baselines. The results obtained by the participants are extensively analyzed to bring interesting insights about the tracks tasks and the nature of the dataset. We wrap up with a discussion on challenge outcomes, and pose several questions that we expect will motivate further scientific research to better understand social cues in human-human and human-machine interaction scenarios and help build future AI applications for good.

KW - AI competitions

KW - Behavior forecasting

KW - Human interaction

KW - Multimodal approaches

KW - Personality recognition

UR - http://www.scopus.com/inward/record.url?scp=85163827521&partnerID=8YFLogxK

M3 - Conference article in Journal

AN - SCOPUS:85163827521

SN - 2640-3498

VL - 173

SP - 4

EP - 52

JO - Proceedings of Machine Learning Research

JF - Proceedings of Machine Learning Research

Y2 - 16 October 2021

ER -

ChaLearn LAP Challenges on Self-Reported Personality Recognition and Non-Verbal Behavior Forecasting During Social Dyadic Interactions: Dataset, Design, and Results

Abstract

Conference

Bibliographical note

Keywords

AUB Link

Other files and links

Fingerprint

Cite this