FaceBots: Steps Towards Enhanced Long-Term Human-Robot Interaction by Utilizing and Publishing Online Social Information
| Autor: | Chandan Datta, Shervin Emami, Andry Tanoto, Michael Petychakis, Chiraz BenAbdelkader, Panos Toulis, Nikolaos Mavridis, Wajahat Kazmi, Alexandros Tsamakos |
|---|---|
| Rok vydání: | 2010 |
| Předmět: |
social networks
Technology Exploit Computer science business.industry Cognitive Neuroscience long-term hri Behavioural sciences conversational robots Social web Human–robot interaction Human-Computer Interaction human-robot interaction Behavioral Neuroscience Developmental Neuroscience Artificial Intelligence Human–computer interaction Systems architecture Robot Onboard camera Artificial intelligence business Natural language face recognition |
| Zdroj: | Paladyn: Journal of Behavioral Robotics, Vol 1, Iss 3, Pp 169-178 (2010) |
| ISSN: | 2081-4836 |
| DOI: | 10.2478/s13230-011-0003-y |
| Popis: | The overarching goal of the FaceBots project is to support the achievement of sustainable long-term human-robot relationships through the creation of robots with face recognition and natural language capabilities, which exploit and publish online information, and especially social information available on Facebook, and which achieve two significant novelties. The underlying experimental hypothesis is that such relationships can be significantly enhanced if the human and the robot are gradually creating a pool of episodic memories that they can co-refer to (“shared memories”), and if they are both embedded in a social web of other humans and robots they mutually know (“shared friends”). We present a description of system architecture, as well as important concrete results regarding face recognition and transferability of training, with training and testing sets coming from either one or a combination of two sources: an onboard camera which can provide sequences of images, as well as facebook-derived photos. Furthermore, early interaction-related results are presented, and evaluation methodologies as well as interesting extensions are discussed. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|