Standard

On Social Involvement in Mingling Scenarios : Detecting Associates of F-formations in Still Images. / Zhang, Lu; Hung, Hayley.

In: IEEE Transactions on Affective Computing, 2019, p. 1-13.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

BibTeX

@article{5dab476fa7ed4a4ab4f9e2bef4c10768,
title = "On Social Involvement in Mingling Scenarios: Detecting Associates of F-formations in Still Images",
abstract = "In this paper, we carry out an extensive study of social involvement in free standing conversing groups (the so-called F-formations) from static images. By introducing a novel feature representation, we show that the standard features which have been used to represent full membership in an F-formation cannot be applied to the detection of associates of F-formations due their sparser occurence. We enrich state-of-the-art F-formation modelling by learning a frustum of attention that accounts for the spatial context. That is, F-formation configurations vary with respect to the arrangement of furniture and the non-uniform crowdednessin the space during mingling scenarios. Moroever, the majority of prior works have considered the labelling of conversing groups as an objective task, requiring only a single annotator. However, we show that by embracing the subjectivity of social involvement, we not only generate a richer model of the social interactions in a scene but can use the detected associates to improve initial estimates of the full members of an F-formation. We carry out extensive experimental validation of our proposed approach by collecting a novel set of multi-annotator labels of involvement on two publicly available datasets; The Idiap Poster Data and SALSA data set.",
keywords = "F-formations Detection, Feature extraction, Human Behaviour Analysis, Labeling, Psychology, Semantics, Social Group Detection, Surveillance, Task analysis, Visualization",
author = "Lu Zhang and Hayley Hung",
note = "This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.",
year = "2019",
doi = "10.1109/TAFFC.2018.2855750",
language = "English",
pages = "1--13",
journal = "IEEE Transactions on Affective Computing",
issn = "1949-3045",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

RIS

TY - JOUR

T1 - On Social Involvement in Mingling Scenarios

T2 - IEEE Transactions on Affective Computing

AU - Zhang, Lu

AU - Hung, Hayley

N1 - This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

PY - 2019

Y1 - 2019

N2 - In this paper, we carry out an extensive study of social involvement in free standing conversing groups (the so-called F-formations) from static images. By introducing a novel feature representation, we show that the standard features which have been used to represent full membership in an F-formation cannot be applied to the detection of associates of F-formations due their sparser occurence. We enrich state-of-the-art F-formation modelling by learning a frustum of attention that accounts for the spatial context. That is, F-formation configurations vary with respect to the arrangement of furniture and the non-uniform crowdednessin the space during mingling scenarios. Moroever, the majority of prior works have considered the labelling of conversing groups as an objective task, requiring only a single annotator. However, we show that by embracing the subjectivity of social involvement, we not only generate a richer model of the social interactions in a scene but can use the detected associates to improve initial estimates of the full members of an F-formation. We carry out extensive experimental validation of our proposed approach by collecting a novel set of multi-annotator labels of involvement on two publicly available datasets; The Idiap Poster Data and SALSA data set.

AB - In this paper, we carry out an extensive study of social involvement in free standing conversing groups (the so-called F-formations) from static images. By introducing a novel feature representation, we show that the standard features which have been used to represent full membership in an F-formation cannot be applied to the detection of associates of F-formations due their sparser occurence. We enrich state-of-the-art F-formation modelling by learning a frustum of attention that accounts for the spatial context. That is, F-formation configurations vary with respect to the arrangement of furniture and the non-uniform crowdednessin the space during mingling scenarios. Moroever, the majority of prior works have considered the labelling of conversing groups as an objective task, requiring only a single annotator. However, we show that by embracing the subjectivity of social involvement, we not only generate a richer model of the social interactions in a scene but can use the detected associates to improve initial estimates of the full members of an F-formation. We carry out extensive experimental validation of our proposed approach by collecting a novel set of multi-annotator labels of involvement on two publicly available datasets; The Idiap Poster Data and SALSA data set.

KW - F-formations Detection

KW - Feature extraction

KW - Human Behaviour Analysis

KW - Labeling

KW - Psychology

KW - Semantics

KW - Social Group Detection

KW - Surveillance

KW - Task analysis

KW - Visualization

UR - http://www.scopus.com/inward/record.url?scp=85050400733&partnerID=8YFLogxK

U2 - 10.1109/TAFFC.2018.2855750

DO - 10.1109/TAFFC.2018.2855750

M3 - Article

SP - 1

EP - 13

JO - IEEE Transactions on Affective Computing

JF - IEEE Transactions on Affective Computing

SN - 1949-3045

ER -

ID: 46669009