The Conflict Escalation Resolution (CONFER) Database

Abstract

Conflict is usually defined as a high level of disagreement taking place when individuals act on incompatible goals, interests, or intentions. Research in human sciences has recognized conflict as one of the main dimensions along which an interaction is perceived and assessed. Hence, automatic estimation of conflict intensity in naturalistic conversations would be a valuable tool for the advancement of human-centered computing and the deployment of novel applications for social skills enhancement including conflict management and negotiation. However, machine analysis of conflict is still limited to just a few works, partially due to an overall lack of suitable annotated data, while it has been mostly approached as a conflict or (dis)agreement detection problem based on audio features only. In this work, we aim to overcome the aforementioned limitations by a) presenting the Conflict Escalation Resolution (CONFER) Database, a set of excerpts from audio-visual recordings of televised political debates where conflicts naturally arise, and b) reporting baseline experiments on audio-visual conflict intensity estimation. The database contains approximately 142 minutes of recordings in Greek language, split over 120 non-overlapping episodes of naturalistic conversations that involve two or three interactants. Subject-and session-independent experiments are conducted on continuous-time (frame-by-frame) estimation of real-valued conflict intensity, as opposed to binary conflict/non-conflict clas- * Corresponding author. E-mail address: christos.georgakis@imperial.ac.uk Preprint submitted to Image and Vision Computing December 20, 2016 A C C E P T E D M A N U S C R I P T ACCEPTED MANUSCRIPTsification. For the problem at hand, the efficiency of various audio and visual features and fusion of them as well as various regression frameworks is examined. Experimental results suggest that there is much room for improvement in the design and development of automated multi-modal approaches to continuous conflict analysis. The CONFER Database is publicly available for non-commercial use at http://ibug.doc.ic.ac.uk/resources/confer/.

SEWA DB: A Rich Database for Audio-Visual Emotion and Sentiment Research in the Wild

“…The gender split is quite skewed with 12 male and only 4 female subjects. The dataset is annotated in terms of Frontchannel (main speaker periods), Backchannel (qualified utterances and expressions), agreement/disagreement episodes, smiles, laughter, negative and positive surprises, thinking phases, confusion, head motions, made using ELAN [50] The Conflict Escalation Resolution (CONFER) [8] dataset is constituted of 120 video clips of interactions between 54 subjects from Greek televised political debates. The data was annotated by 10 experts in terms of continuous conflict intensity.…”

Online Attention for Interpretable Conflict Estimation in Political Debates

“…We make use of the Conflict Escalation Resolution (CON-FER) database [17], a set of recordings of televised political debates from Greek TV. It contains 73 episodes of dyadic interactions and 47 episodes of interactions among three subjects, spanning approximately 142 minutes with 54 subjects.…”

“…All our experiments follow the 5-fold cross-validation experimental protocol as proposed by [17] to represent a fair and representative view over the whole dataset. This means that for each fold the model is trained on 3 out of 5 parts of the dataset, validation is performed using a 4th and test metrics are calculated over a 5th.…”

“…We transform these with an affine transformation based on 5 stable points: those corresponding to the corners of the eyes and the tip of the nose. Then we extract the following features: 1) Expr.+SIFT: We reuse the best-performing feature set reported by [17] to allow for a fair comparison. We only consider the Expression and SIFT feature sets in conjunction (and not separate), as [17] reports this makes up their bestperforming visual feature set.…”

“…To this end, we propose an online attention-based neural network that is learned end-toend from facial and vocal features. We evaluate the proposed model in conflict estimation in political debates by conducting experiments on the CONFER dataset [17]. Significant improvements over the state of the art are reported.…”