Drug companies have been told to publish their funding of patients' groups. Andrew Jack says ... more Drug companies have been told to publish their funding of patients' groups. Andrew Jack says that these groups should open their books too
Multimodal intelligent information presentation, 2005
Document production is an important function in many organizations. In addition to products such ... more Document production is an important function in many organizations. In addition to products such as technical papers, instruction manuals, system documentation, courseware, etc., briefings are a very common type of multimedia document product. Since so much time is spent by so many people in producing briefings, often under serious time constraints, any method to reduce the amount of time spent on briefing production could yield great gains in productivity.
The main goal of the study at hand is the development of a multimodal, ensemble based system for ... more The main goal of the study at hand is the development of a multimodal, ensemble based system for emotion recognition. Special attention is given to a problem often neglected: the problem of missing data. In off-line evaluation the problem can be easily ...
I met Jeff Rickel for the first time at one of the first workshops on Lifelike Agents in Japan in... more I met Jeff Rickel for the first time at one of the first workshops on Lifelike Agents in Japan in 1997. Jeff presented Steve, an animated pedagogical agent for training. I was fascinated by Jeff’s idea to create an agent that inhabits a 3D environment with students enabling completely new forms of interaction. While the behaviour of most animated agents at that time was scripted, Jeff succeeded in realizing an agent that autonomously plans its behaviours and interacts with a student using gestures and natural language. Instead of relying on pre-designed animation sequences and audio clips, the 3D character Steve could be directly controlled by commands, such as “look at“, “walk to“ or “grasp an object“. Another interesting feature of the approach was Steve’s ability to dynamically switch between coaching and demonstration activities. I was deeply impressed about the advanced technology used in Steve and surprised that I had not heard about this amazing work before. Only a short time later, Steve has become ubiquitous in papers on lifelike animated agents and inspired a number of related projects world-wide. Since 1997, Jeff has been continuously working on Steve’s interaction and pedagogical capabilities. An interesting extension is the support of team tasks by the creation of multiple instances of Steve agents that may serve as instructors or substitute missing team members. The basic idea has been published in a seminal paper with Lewis Johnson at the 9th International Conference on AI in Education (AI-ED 1999) and was honoured with an Outstanding Paper Award. While many Artificial Intelligence approaches today still focus on toy examples, Steve represents a complex real-world application integrating technologies from various areas of Artificial Intelligence, such as planning, natural language processing, knowledge-based graphics and animation, agent-based architectures and multimodal interfaces. The high potential of the Steve work has also been recognized by the IJCAI Distinguished Paper Track Committee who invited Jeff and Lewis to publish an extended version of their AI-ED 1999 paper in the prestigious volume “Exploring Artificial Intelligence in the New Millennium“ which features award-winning papers from Artificial Intelligence. I believe the creation of agents that can take on different roles is a major accomplishment of Jeff’s work which has been highly influential for later work on interactive storytelling. An example includes the ambitious MRE project to which Jeff has been serving as a major contributor. The objective of this project is the creation of a compelling and engaging tool for training in peacekeeping situations by the integration of virtual reality with Hollywood storytelling techniques (see also the article on MRE in this Special Issue). While agent-based interfaces usually focus on dialogue between a single human and a single agent, the MRE project aims at face-to-face communication among a group of people and synthetic agents. Jeff’s paper with David Traum at the 1st International Joint Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2002) is among the first to meet this challenge. A significant result of the paper is a general model for multi-party conversation based on the Steve architecture and prior work on spoken dialogue. This work opens the door to a completely new class of dialogue systems with much higher resemblance to conversation between humans than has ever been addressed before. In parallel with his work on ICT’s MRE project, Jeff started a fruitful cooperation with colleagues from MERL. Even though there are obvious connections between intelligent tutoring systems and collaborative dialogue, the two areas have been handled separately so far. With the joint development of the PACO agent, this gap has been closed. An important outcome of Jeff’s cooperation with MERL was the idea to model tutorial behaviours as collaborative discourse acts. Jeff has been highly recognized both as a researcher and a person. Last year, he visited my lab in Augsburg where we had a wonderful time together. Since I use to present Steve as a classical example of an AI-based interface agent in my lectures, my students knew about Jeff’s fundamental achievements in the field. In the beginning, they were obviously expecting a more reserved behaviour of a distinguished researcher like him. Therefore, they were rather surprised when they realized how easy it was to get in touch with him and delighted that he spent so much time on discussing their ideas with them. Jeff has done pioneering work on embodied conversational agents. He has provided valuable inspirations for a number of related areas, such as agent-based architectures, tutoring systems, computer graphics, natural language dialogue and affective user modelling. His ideas have influenced the research of many of us. Even though he was struggling with cancer, he has been working up to the end. He was actively involved with…
Proceedings of the Workshop on Human-Habitat for Health (H3): Human-Habitat Multimodal Interaction for Promoting Health and Well-Being in the Internet of Things Era
Proceedings of the Workshop on Human-Habitat for Health (H3): Human-Habitat Multimodal Interaction for Promoting Health and Well-Being in the Internet of Things Era
Drug companies have been told to publish their funding of patients' groups. Andrew Jack says ... more Drug companies have been told to publish their funding of patients' groups. Andrew Jack says that these groups should open their books too
Multimodal intelligent information presentation, 2005
Document production is an important function in many organizations. In addition to products such ... more Document production is an important function in many organizations. In addition to products such as technical papers, instruction manuals, system documentation, courseware, etc., briefings are a very common type of multimedia document product. Since so much time is spent by so many people in producing briefings, often under serious time constraints, any method to reduce the amount of time spent on briefing production could yield great gains in productivity.
The main goal of the study at hand is the development of a multimodal, ensemble based system for ... more The main goal of the study at hand is the development of a multimodal, ensemble based system for emotion recognition. Special attention is given to a problem often neglected: the problem of missing data. In off-line evaluation the problem can be easily ...
I met Jeff Rickel for the first time at one of the first workshops on Lifelike Agents in Japan in... more I met Jeff Rickel for the first time at one of the first workshops on Lifelike Agents in Japan in 1997. Jeff presented Steve, an animated pedagogical agent for training. I was fascinated by Jeff’s idea to create an agent that inhabits a 3D environment with students enabling completely new forms of interaction. While the behaviour of most animated agents at that time was scripted, Jeff succeeded in realizing an agent that autonomously plans its behaviours and interacts with a student using gestures and natural language. Instead of relying on pre-designed animation sequences and audio clips, the 3D character Steve could be directly controlled by commands, such as “look at“, “walk to“ or “grasp an object“. Another interesting feature of the approach was Steve’s ability to dynamically switch between coaching and demonstration activities. I was deeply impressed about the advanced technology used in Steve and surprised that I had not heard about this amazing work before. Only a short time later, Steve has become ubiquitous in papers on lifelike animated agents and inspired a number of related projects world-wide. Since 1997, Jeff has been continuously working on Steve’s interaction and pedagogical capabilities. An interesting extension is the support of team tasks by the creation of multiple instances of Steve agents that may serve as instructors or substitute missing team members. The basic idea has been published in a seminal paper with Lewis Johnson at the 9th International Conference on AI in Education (AI-ED 1999) and was honoured with an Outstanding Paper Award. While many Artificial Intelligence approaches today still focus on toy examples, Steve represents a complex real-world application integrating technologies from various areas of Artificial Intelligence, such as planning, natural language processing, knowledge-based graphics and animation, agent-based architectures and multimodal interfaces. The high potential of the Steve work has also been recognized by the IJCAI Distinguished Paper Track Committee who invited Jeff and Lewis to publish an extended version of their AI-ED 1999 paper in the prestigious volume “Exploring Artificial Intelligence in the New Millennium“ which features award-winning papers from Artificial Intelligence. I believe the creation of agents that can take on different roles is a major accomplishment of Jeff’s work which has been highly influential for later work on interactive storytelling. An example includes the ambitious MRE project to which Jeff has been serving as a major contributor. The objective of this project is the creation of a compelling and engaging tool for training in peacekeeping situations by the integration of virtual reality with Hollywood storytelling techniques (see also the article on MRE in this Special Issue). While agent-based interfaces usually focus on dialogue between a single human and a single agent, the MRE project aims at face-to-face communication among a group of people and synthetic agents. Jeff’s paper with David Traum at the 1st International Joint Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2002) is among the first to meet this challenge. A significant result of the paper is a general model for multi-party conversation based on the Steve architecture and prior work on spoken dialogue. This work opens the door to a completely new class of dialogue systems with much higher resemblance to conversation between humans than has ever been addressed before. In parallel with his work on ICT’s MRE project, Jeff started a fruitful cooperation with colleagues from MERL. Even though there are obvious connections between intelligent tutoring systems and collaborative dialogue, the two areas have been handled separately so far. With the joint development of the PACO agent, this gap has been closed. An important outcome of Jeff’s cooperation with MERL was the idea to model tutorial behaviours as collaborative discourse acts. Jeff has been highly recognized both as a researcher and a person. Last year, he visited my lab in Augsburg where we had a wonderful time together. Since I use to present Steve as a classical example of an AI-based interface agent in my lectures, my students knew about Jeff’s fundamental achievements in the field. In the beginning, they were obviously expecting a more reserved behaviour of a distinguished researcher like him. Therefore, they were rather surprised when they realized how easy it was to get in touch with him and delighted that he spent so much time on discussing their ideas with them. Jeff has done pioneering work on embodied conversational agents. He has provided valuable inspirations for a number of related areas, such as agent-based architectures, tutoring systems, computer graphics, natural language dialogue and affective user modelling. His ideas have influenced the research of many of us. Even though he was struggling with cancer, he has been working up to the end. He was actively involved with…
Proceedings of the Workshop on Human-Habitat for Health (H3): Human-Habitat Multimodal Interaction for Promoting Health and Well-Being in the Internet of Things Era
Proceedings of the Workshop on Human-Habitat for Health (H3): Human-Habitat Multimodal Interaction for Promoting Health and Well-Being in the Internet of Things Era
Uploads