Final Report Summary - CV-SUPER (Computer Vision for Scene Understanding from a first-person Perspective)
The goal of CV-SUPER has been to create the technology to perform dynamic visual scene understanding from the perspective of a moving human observer. Briefly stated, we want to enable computers to see and understand what humans see when they navigate their way through busy inner-city locations. Our target is dynamic visual scene understanding in public spaces, such as in street traffic, in pedestrian zones, in shopping malls, or in other locations primarily built for humans. Within CV-SUPER, computer vision algorithms were developed that can recognize other traffic participants in such settings, interpret and understand their actions and their interactions with other people and inanimate objects, and from this understanding derive predictions of their future behaviors within the next few seconds.
In order to reach this ambitious goal, the project focused on addressing several principled limitations of previous dynamic scene understanding approaches. Among the main technical achievements are the development of novel and scalable approaches for detecting and tracking people and analyzing their interactions; for robustly tracking generic unknown objects; and for inferring the semantic properties of the observed environment. Those components will play an important role for the creation of technical systems that may one day assist humans in their daily lives within busy public spaces, e.g. in the form of personal assistance devices for elderly or visually impaired people or in the form of future generations of mobile service robots and intelligent vehicles.
In order to reach this ambitious goal, the project focused on addressing several principled limitations of previous dynamic scene understanding approaches. Among the main technical achievements are the development of novel and scalable approaches for detecting and tracking people and analyzing their interactions; for robustly tracking generic unknown objects; and for inferring the semantic properties of the observed environment. Those components will play an important role for the creation of technical systems that may one day assist humans in their daily lives within busy public spaces, e.g. in the form of personal assistance devices for elderly or visually impaired people or in the form of future generations of mobile service robots and intelligent vehicles.