The Visual Social Distancing Problem

One of the main and most effective measures to contain the recent viral outbreak is themaintenance of the so-called Social Distancing (SD). To comply with this constraint, governments areadopting restrictions over the minimum inter-personal distance between people. Given this actual scenario,it is crucial to massively measure the compliance to such physical constraint in our life, in order to figure outthe reasons of the possible breaks of such distance limitations, and understand if this implies a potential threat.To this end, we introduce the Visual Social Distancing (VSD) problem, defined as the automatic estimationof the inter-personal distance from an image, and the characterization of related people aggregations. VSDis pivotal for a non-invasive analysis to whether people comply with the SD restriction, and to providestatistics about the level of safety of specific areas whenever this constraint is violated. We first point outthat measuring VSD is not only a geometrical problem, but it also implies a deeper understanding of thesocial behaviour in the scene. The aim is to truly detect potentially dangerous situations while avoiding falsealarms (e.g., a family with children or relatives, an elder with their caregivers), all of this by complyingwith current privacy policies. We then discuss how VSD relates with previous literature in Social SignalProcessing and indicate a path to research new Computer Vision methods that can possibly provide a solutionto such problem. We conclude with future challenges related to the effectiveness of VSD systems, ethical implications and future application scenarios.