Imaging and Multimedia
This research area includes computer vision, image processing, production, management and transmission of multimedia content and human-machine interaction.
Computer Vision allows the computer to identify salient elements in a scene (such as objects, people or actions) and take appropriate actions depending on the specific purpose, as for human vision. The connections of this discipline with artificial intelligence are very tight, as the computer has to interpret the acquired images. The aim is therefore not only to see, but also to elaborate and provide useful results based on observation. For example, if an obstacle suddenly appears before us while driving, in a very short time our visual system identifies the main object of the scene (the obstacle) and sends the brain stimuli to decide what to do (avoid it). The goal of computer vision is to allow computers to perform the same type of activity with the highest possible efficiency.
Image processing is closely linked to machine vision, since it is always necessary for the extraction of information on the content of the acquired images. However, the techniques can be applied to images of any type, not necessarily resulting from real-time vision processes. This research area includes image enhancement and restoration (for example in the medical or remote sensing context), scene segmentation, and Pattern Recognition, as well as structural descriptions and representations of knowledge for learning. Machine and Deep Learning are increasingly used in this field, where they are particularly effective. 3D modeling techniques can be used to simulate objects or environments in different contexts (such as Digital Humanities).
Multimedia content, especially for the Web, has now assumed a fundamental role in communication, and the correct choice, production, management and transmission are indispensable factors for the success of a communicative project.
Multimedia is often considered in association with the interaction between the user and the machine. For example, a website rich in content that is not easily and intuitively reachable is of little use. For effective communication it is therefore important that the interface between man and machine is usable. In addition to traditional modes of communication, alternative solutions can increasingly be found, such as those based on perceptual interfaces, which provide the computer with typically human sensory abilities. Examples are gestural interfaces and interfaces based on eye tracking, the latter being particularly useful as assistive technologies. The application contexts are not only those of explicit communication with the machine, but also transversal areas such as the study of user behavior and biometrics.
Click on the link to discover the contribution and the ongoing research in the specific laboratory on this research topic: