Giulia Paggetti

Social interaction discovery by statistical analysis of F-formations.

We present a novel approach for detecting social interactions in a crowded scene by employing solely visual cues. The detection of social interactions in unconstrained scenarios is a valuable and important task, especially for surveillance purposes. Our proposal is inspired by the social signaling literature, and in particular it considers the sociological notion of F-formation. An F-formation is a set of possible configurations in space that people may assume while participating in a social interaction. Our system takes as input the positions of the people in a scene and their (head) orientations; then, employing a voting strategy based on the Hough transform, it recognizes F-formations and the individuals associated with them. Experiments on simulations and real data promote our idea.

Social interactions by visual focus of attention in a three-dimensional environment

In human behaviour analysis, the visual focus of attention (VFOA) of a person is a very important cue. VFOA detection is difficult, though, especially in a unconstrained and crowded environment, typical of video surveillance scenarios. In this paper, we estimate the VFOA by defining the Subjective View Frustum, which approximates the visual field of a person in a three-dimensional representation of the scene. This opens up to several intriguing behavioural investigations. In particular, we propose the Inter-Relation Pattern Matrix, which suggests possible social interactions between the people present in a scene. Theoretical justifications and experimental results substantiate the validity and the goodness of the analysis performed.

Towards Computational Proxemics: Inferring Social Relations from Interpersonal Distances

This paper proposes a study corroborated by preliminary experiments on the inference of social relations based on the analysis of interpersonal distances, measured with on obtrusive computer vision techniques. The experiments have been performed over 13 individuals involved in casual standing conversations and the results show that people tend to get closer when their relation is more intimate. In other words, social and physical distances tend to match one another. In this respect, the results match the findings of proxemics, the discipline studying the social and affective meaning of space use and organization in social gatherings. The match between results and expectations of proxemics is observed also when changing one of the most important contextual factors in this type of scenarios, namely the amount of space available to the interactants.

Re-locating colors in the OSA space

A monolexemic color-naming experiment in the Italian language was performed on a sample of 1,014 colors obtained extending the OSA-UCS color system (E-OSA). Colors were rendered on a Mitsubishi Diamond pro 2070 monitor as patches subtending 10° of visual angle in a completely dark room. The 11 universal categories proposed by Berlin and Kay (Basic color terms: Their universality and evolution, 1969) were considered. Consensus, consistency and focal colors were determined for each category. Comparisons were performed among the results obtained using Boynton & Olson (B&O), the extended OSA (E-OSA), a subset of it covering almost all the B&O sample (R-OSA) and the Sturges & Whitfield (S&W) sample sets. A good overlap could be observed among the locations of the consensus colors in the {L, j, g} color model between B&O and R-OSA, as well as a close proximity among the centroids of homologue regions for the majority of the classes. The analysis of the location of focal colors versus the centroids suggests that an additional basic term could be missing for the Italian language in the blue category.

Analyzing Groups: A Social Signaling Perspective

This chapter introduces some basic methods to deal with groups of people in surveillance settings. Recently, modeling groups has become a very active trend for video surveillance researchers. Our solution is proper of the recently forged field of social signaling, since it embeds notions of social psychology into computer vision techniques, offering a novel research perspective for the video surveillance community. In particular, we present methods to discover and track groups of people, and to infer what is the focus of attention of each person, that is, we estimate the portion of a scene that is frequently observed by people. Each method we present is evaluated in an experimental section on real scenario, that gives a clear idea of its performance and potentialities.

The role of the posterior parietal cortex in stereopsis and hand-eye coordination during motor task behaviours

The field of ‘Neuroergonomics’ has the potential to improve safety in high-risk operative environments through a better appreciation of the way in which the brain responds during human–tool interactions. This is especially relevant to minimally invasive surgery (MIS). Amongst the many challenges imposed on the surgeon by traditional MIS (laparoscopy), arguably the greatest is the loss of depth perception. Robotic MIS platforms, on the other hand, provide the surgeon with a magnified three-dimensional view of the environment, and as a result may offload a degree of the cognitive burden. The posterior parietal cortex (PPC) plays an integral role in human depth perception. Therefore, it can be hypothesized that differences in PPC activation between monoscopic and stereoscopic vision may be observed. In order to investigate this hypothesis, the current study explores disparities in PPC responses between monoscopic and stereoscopic visual perception to better de-couple the burden imposed by laparoscopy and robotic surgery on the operator’s brain. Fourteen participants conducted tasks of depth perception and hand-eye coordination under both monoscopic and stereoscopic visual feedback. Cortical haemodynamic responses were monitored throughout using optical functional neuroimaging. Overall, recruitment of the bilateral superior parietal lobule was observed during both depth perception and hand-eye coordination tasks. This occurred contrary to our hypothesis, regardless of the mode of visual feedback. Operator technical performance was significantly different in two- and three-dimensional visual displays. These differences in technical performance do not appear to be explained by significant differences in parietal lobe processing.