Facial expressions have been deeply studied and are often the primary source of information related to a person who is being observed. But what changes when a computer observes humans?
The human eye is a perfect machine for analyzing images from a technical, semiotic and emotional point of view, as it is the perfect tool for analyzing body language.
Body language represents first-level information captured by the sense of sight and then translated into rational judgment or emotion by the brain. Let’s try to replace eyes and brain with the corresponding digital technology and we will have a taste of the near future at its bright and dark side.
When we look at an image we can assign it to a category, when we look at a gesture or an expression we can assign a meaning. Sometimes the meaning of a gesture represents a cultural sediment but it can be traced back to a language that is practiced daily and that has now been codified as universal also thanks to digital and media.
The technical analysis of images has long been the domain of machines, but gestures and body movements have become a central point of research for computer vision.
Recognizing the body language through a computer, attributing a meaning to a specific movement means to be able to put in relation and communication the man with the computer, an instance able to process the information, aggregate them, analyze them in real time and manage them to elaborate feedback, specific actions, responses and information.
Re-shaping Prossemic zones with computer vision
Computer vision has also managed in some way to reshape the so-called proxemic zones. To give a few examples, what is defined as an intimate area, closeness, can correspond, in the relationship between man and device, to the recognition and the possibility of unlocking an action or a content. When a proximity relationship occurs in this area we are likely faced with a real interest or a relationship between man and object that can be interpreted as “ownership”.
The intimate space, personal space, the social space, and public space become more and more matter of analysis especially in this recent scenario dominated by the fear of pandemic. People tracking and social distancing measurement are now a prerequisite for public safety and health.
If before the personal and social spaces were to be understood as the distance that we "choose" to put between us and others, now these spaces often become subject to regulation by machines that can technically analyze images.
Undoubtedly if we look at this reality from a perspective marked by the pandemic emergency, what follows is a dystopian reality worthy of the apocalyptic visions of Huxley or Orwell.
If instead we adopt a perspective guided by the light of innovation, the scenario can be the Utopian one in which the processing and technical analysis of images allows to reach high safety standards and high quality standards in services, we think for example of applications of computer vision in the automotive sector, home automation, transport, health and also education.
Natural User Interface and Touchless User Interface
Body recognition, voice, gestures and position detection allow the development of natural interfaces that cancel out any technological interference in the experience.
This leads us to open an in-depth study on the concept of Natural User Interface and Touchless User Interface.
A gesture is a movement of the body that contains information and this is the premise of human language (and animal language in the broad sense) but on this premise is also based the design of services that take an approach that focuses on the Natural User Interface.
The prerogative of this approach is to allow a "seamless" interaction between man and machine by making the interface disappear and leaving only the experience as a form of content to surface.
The recognition of the body and gestures can be based on the use of cameras, sensors or other devices capable of detecting data to which a value is attributed.
This technical analysis of the images or of the movement carried out naturally by the human eye, can be carried out by the machines according to two distinct approaches. An approach is based on the collection of images, which compared to a comparison database allows to attribute a given and a meaning to the image collected. The second approach is based on the identification of the skeleton and therefore is faster and more precise because it is based on points and coordinates that correspond to a gestural encoding well recognizable. This approach is in fact adopted by the gaming industry that can not disregard the speed of response, precision and allows you to capture and interpret a wide range of gestures.
The Touchless User Interface (TUI) is based on the premise that, with the use of cameras, 3D sensors and computer vision, there is an understanding of movement or gestural or vocal actions. This means that no device contact is required but a gesture or voice input can be enough to activate a feedback response of any kind.
Human behavior recognition techniques can be used variously to improve services, make them more accessible and certainly safer from a health point of view.
Learn how using space sensors can improve queue management
SUBSCRIBE TO OUR NEWSLETTER
Mon - Fri: 10am - 18pm