Virtual Human Interface Group
 |
| Our Mission |
|
To mimic the quality of everyday human communication, future computer interfaces must combine the benefits of high visual fidelity animated human agents with conversational intelligence and the ability to modulate the emotions of their users in a personalized manner. During the past several decades, researchers have conducted countless studies on agents and human animation in order to create a HCI that works by utilizing the natural means of interaction, e.g. words, gestures, glances and body language instead of traditional computer devices, such as the keyboard and mouse.
From a computer science and interface design perspective to achieve the above requires the ability to create believable digital humans with photo-real faces capable of expressing the finest shades of emotions in a controllable manner. The same system must also be able to read the users' emotional reactions and adapt the behavior of the digital human accordingly in real-time. The group's research is focused on implementing this concept as a step towards creating a novel face-to-face HCI solution called the Virtual Human Interface. Our solution builds upon many years of interdisciplinary research to create a closed-loop model of interaction whereas the user's internal state (emotion, level of attention, etc.) is constantly monitored and driven directly by the animated character with the purpose of creating emotional bonding. This emotional bonding then acts as a catalyst to help turning information into knowledge. In other words, our advanced user interface draws on emotions to help its users (most frequently students) in the learning process by intelligently tailoring its workload and constantly adapting its presentation strategies, hence the name affective intelligence. It is therefore argued that photo-realistic virtual humans due to their similarities to their real-life counterparts make a powerful affective interface and as such they will likely and quickly become the primary means of communication between computers and humans for future decades to come.
What possibly hindered computer scientists for decades in achieving the above goals introduced above is the failure to recognize that this level of computer generated intelligence and its forms of expression lie on the very border of science, art and entertainment. While traditional research fields such as Artificial Intelligence (AI), cognitive architectures, user modeling, ambient intelligence, information-oriented human dialogue design and the various information exchange protocols to implement them as well as the tools of computer graphics, computer vision, image processing, pattern recognition in general and face recognition in particular, and finally computer linguistics are only required perquisites despite each having grown into full-fledged areas of scientific interest and specialization. To overcome these shortcomings we designed and implemented a seamless user interface where animation and perception is blended into a single and unified framework built upon concise models of human face-to-face communication.
One of the key points of the VHI group's research is the workings of our psychology and how the brain works when it comes to processing intention cues, emotions and other means (commonly called non-verbal signs) has to be included in the model of the interaction process. Specifically, since facial information processing plays a vital role in our evolution and thus it represents a dedicated subsystem in our visual system, The acute capability of people to read various cues from faces offers the unique opportunity to use high fidelity digital faces as a prime interface. Such a communication layer provides thousands of parallel channels readily understandable even by the least educated or youngest members of our society. Recent advances in computer graphics in general, and virtual humans in particular have laid the foundation to be able to achieve the photo-realistic rendering such an interface requires. However it is not as much the visual appearance that holds the key to a successful implementation but the precise timing of facial displays expressing a coherent message in yet to be understood language of non-verbal signs governed by the rules of turn-taking. It is precisely those emotions, modulated by the novel computer interface we created, that establish the foundation to open an extremely powerful information gate and throughput to our brains.
|
Head of department
- Barnabás Takács, Ph.D.
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: btakacs@sztaki.hu
Staff
- Balázs Benedek (architecture and sensory systems)
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: bbenedek@sztaki.hu
- Gábor Szijártó (computer graphics, real-time shaders & effects)
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: szijarto@sztaki.hu
- Dávid Hanák (artificial intelligence, behaviour & emotions)
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: dhanak@sztaki.hu
- Levente Dobson (computer vision & image understanding)
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: dobson@sztaki.hu
- Alex Beregszászi (Multi-media architectures)
- Gergely Mészáros-Komáromy (Mobile platforms)
Address: 1111 Budapest, Kende u. 13-17.Room number: K 307Phone: +36 1 279 6183E-mail: komaromy@sztaki.hu
Departments 
Magyar
