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.

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.

No abstract

The present study investigated Italian basic color terms (BCTs). It is an extension of our previous work that explored Italian basic color categories (BCCs) using a constrained colornaming method, with 11 Italian BCTs allowed, including blu for naming the BLUE area. Since a latter outcome indicated a categorization bias, here monolexemic color-naming method was employed, enabling also use of azzurro, deeply entrenched Italian term that designates light blue. In Experiment 1, colors (N=367), sampling the Munsell Mercator projection, were presented on a CRT; color names and reaction times of vocalization onset were recorded. Naming consistency and consensus were estimated. Consistency was obtained for 12 CTs, including the two blue terms; consensus was found for 11 CTs, excluding rosso 'red'. For each consensus category, color with the shortest RT was considered focal. In Experiment 2, consensus stimuli (N=72) were presented; on each trial, observers indicated the focal color ("best example") in an array of colors comprising a consensus category. For each of the 12 Italian CCs, centroid was calculated and focal color (two measures) estimated. Compared to English color terms, two outcomes are specific to Italian color naming: (i) naming of the RED-PURPLE area is highly refined, with consistent use of emergent non-BCTs; (ii) azzurro and blu both perform as BCTs dividing the BLUE area along the lightness dimension. The findings are considered in the framework of the weak relativity hypothesis. Historico-linguistic, environmental and pragmatic communication factors are discussed that conceivably have driven the extension of the BCT inventory in Italian.

In a previous work, the authors reported on the results of a color naming experiment performed on native Italian speakers regarding the location of focal colors and centroids in the Uniform Color Scales of the Optical Society of America color system. That work was aiming at comparing such data with those previously obtained by Boynton and Olson (B&O) accounting for the differences in the paradigm and the language. The number of consistency and consensus colors in the different lightness plans was also reported but no information was provided on their placement. Though, such information is very important for any subsequent modeling stage. The objective of this article is to fill such a gap and share such data with the scientific community to provide a reference database for future investigation. Three different datasets were considered: the extended OSA (E‐OSA), the reduced OSA (R‐OSA), and the B&O's (B&O) sets of reference colors. Results show a good overlap among the locations of the consensus colors in the {L, j, g} color model between B&O and the subset of E‐OSA colors overlapping with the B&O 424 colors (R‐OSA), as well as a strong agreement on consistency. Furthermore, a close proximity among the centroids of homologue regions for the majority of the classes was found. © 2012 Wiley Periodicals, Inc. Col Res Appl, 38, 437–447, 2013