Citation Context

...an increasingly important research area. It involves computer vision, machine learning and behavioral sciences, and can be used for many applications such as security [20], human-computer-interaction =-=[23]-=-, driver safety [24], and health-care [17]. Significant advances have been made in the field over the past decade [21, 22, 27] with increasing interest in non-posed facial behavior in naturalistic con...

Citation Context

...ncludes 70 debates for a total of 43 hours and 10 minutes of material. 1. Introduction As automatic analysis of social interactions attracts increasingly more attention in the computing community [3] =-=[8]-=-, publicly available benchmarks become a crucial element for the progress of the domain. Benchmarks allow different researchers to apply the same experimental protocols over the same data and this is ...

Citation Context

...ion [13]. That is naturally the case not only for agreement and disagreement, but for all facets of human social behavior, including politeness, flirting, social relations, and other social attitudes =-=[78]-=-. Machine analysis of nonverbal behavioral cues (e.g. blinks, smiles, nods, crossed arms, etc.), have recently been the focus of intensive research, as surveyed by Pantic et al. in [56, 58]. Similarly...

Citation Context

...techniques, often relying on mining speech or facial expressions. 2.2.1. Recognizing Emotions from Voice and Video Existing work in affect recognition focuses primarily on the recognition of emotions =-=[36, 39]-=-; and does so largely by leveraging the visual [7, 10, 27] and acoustic [20, 32] signals found in the speech, actions and faces of people. While authors sometimes use emotion and mood interchangeably,...

Citation Context

... Broadcast Data, Meeting Recordings. I. INTRODUCTION The computing community is making significant efforts towards the development of automatic approaches for the analysis of social interactions (see =-=[1]-=-[2][3] for extensive surveys of the domain). This is not surprising as social interactions are not only one of the most important aspects of our everyday lives, but also an ubiquitous subject in multi...

Citation Context

...g that nonverbal vocal behavior significantly influences personality perception [9] (see Section 4 for a short survey). Furthermore, domains like affective computing [10] and social signal processing =-=[11]-=- have shown that nonverbal behavioral cues (e.g., prosody, facial expressions, gestures, postures, etc.) work effectively as evidence for technologies dealing with emotional and social phenomena. IEEE...

Citation Context

...and also to deal with the more complex state representations that will be required by the updated high-level reasoning system. To address this, we will draw on recent work in social signal processing =-=[35]-=- by training supervised learning classifiers on data gathered from humans interacting with both real and artificial bartenders, using methods similar to those employed by [6, 18]. 2.4 High-level plann...

Citation Context

...ve processes like attention (interest) and boredom and attitudinal states likes(dis)liking and (dis)agreement, very few efforts towards automatic recognition of these states have beensreported so far =-=[16]-=-. Arguably the most advanced method proposed up to date for detecting the level ofsinterest is that by Schuller et al. [13], who applied Support Vector Regression on a large number ofsfeatures extract...