Introduction to Nagao Lab.
1. Lab Slogan
Nagao Lab. has started researches since April, 2002. Our activities are diverse.
I have a strong philosophy that it is the most important for both undergraduates and graduates to participate in the laboratory activities on their own will and to learn through excogitative creation and practical experiences. So it might be said that the motto of this lab is "Do what you want to do in your way, but on the condition that the way should be meaningful."
However, the students have given to me an impression that they do not seem to know what they really want to do. Or they do not seem to have anything about what they want to do in the first place. This negative impression frustrated my plan to create an ideal laboratory already. In short, quite a few students advanced to the next year before mastering academic work.
It is obvious, even for a new comer with no experience of teaching like me, that university reform cannot go smoothly. Still I has a strong hope of altering the student's state of consciousness little by little through lab activities. I think that there is no sense in attending graduate school without the ability and will of becoming involved in laboratory activities. I also feel that even though the student did his research (or research-like stuff), his experience would not satisfy him.
For this reason, there are some points to be emphasized by me. One is the importance of choosing the research theme after sufficient consultation. Another is that the members and I should discuss their own subjects, not only introducing other papers but accompanying their own ideas and related research. In order to position the significance of each problem to an entire one for the lab, I am planning to make all the members to share the result of their research. For instance, when someone creates a software system, others can use or evaluate it.
2. Covering Research Themes
Although I expect the students to research independently, some may have difficulty in determining a subject of research. For such students, I have set up some following tentative themes. They are, needless to say, themes that I have engaged in from past to present.
Natural Language Processing
To research and develop methods to understand and generate human languages
Examples: Document Processing, Automated Summarization and Translation, Spoken Dialogue Systems
To research and develop methods to integrate information content with real world objects and environments
Examples: Augmented Library, Real World Navigation Systems
To research and develop user friendly human interfaces
Examples: Multimodal Interfaces, Real World Oriented Interfaces, Community Support Systems
To research and develop autonomous systems that can communicate with humans and other autonomous systems such as software and physical robots
Examples: Personified Agent, Social Agent, Conversational Robot
Digital Content Technologies
To research and develop tools and applications to process and generate online content such as digital documents
Examples: XML Database, Annotation, Transcoding, Knowledge Discovery
To research and develop methods to analyze, retrieve, and manipulate multimedia content
Examples: Video Annotation, Video Transcoding, Multimedia Content Retrieval
I think that these research themes can be good guidelines to set students' starting points and goals since all of them are application-oriented and ordinary people can judge how good results we made based on these studies. One of my policies is that we, especially researchers in engineering fields, have to clarify how useful and significant our created theories and technologies are in our future lives.
3. Ongoing Research Projects
Currently the following research activities have been doing in Nagao lab. Of course since the lab only just started, we have repeated the process of trial and error. The direction and style of research have not yet been established. All the more reason, the students would make a lot of discoveries and would think it worthwhile.
Multimedia Annotation and its Applications
This is a support system of creating meta-information for multimedia content and a research on retrieval, summarization and translation of the content through the system. As a broadband network develops, it seems that the distribution of multimedia content becomes more active. Today's reality of our depending much on web search-engine shows that such technology is expected to be just as imperative as search-engine or more than it. Multimedia annotation includes all the information such as video scene structure, audio transcripts and their linguistic structure. This can realize semantic search and summarization for video with higher accuracy than with traditional technology.
We are developing a system that produces minutes for such as meeting semi-automatically. Then, not only we search for content of minutes from the information of date and attendants, but also we have been studying on the method of searching for a similar discussion to an ongoing discussion from the past ones, and on the method of producing an answer to a certain question based on the content of the minutes. In terms of mailing list and online discussion on bulletin board, various studies have been done. However, what we think is greatly different from them including actual meetings. We analyze meetings from diversified perspectives using audio and visual information. We can call this not just data mining but a kind of real-world human activity mining.
Musical Annotation and its Applications
Various methods of describing music electronically have been developed. MIDI is one of them and it is a system of sounding a digital sound resource, not a system of describing general information related music. Recently MusicXML, which describes music content such as score, lyrics and sound source information, was introduced by the use of XML. In this lab, we have done a research to expand MusicXML further with annotation and a research on variety of applications with the use of it, especially on music summarization.
AT, a Personal Intelligent Vehicle: Mountable Computing and Integration of Physical/Informational Environments
Extending the concept of mobile or wearable computing opens up a new field of human-centered computing called Mountable Computing, which supports not only anytime/anywhere computing functionalities but also computer-augmented physical movement and transportation activities. The so-called ITS (Intelligent Transportation System) aims at the augmentation of a conventional traffic system by applying information technologies such as wireless data communication and global positioning systems to location/route guidance, toll gate transactions, and traffic monitoring. While the current ITS does not deal with services for small personal vehicles (motorcycles, electric wheelchairs, and bicycles), we have been studying physical and information infrastructures in order to make such small vehicles intelligent enough to contribute to a physical/digital barrier-free society. We also designed and developed a new electric personal vehicle called the AT. AT stands for Attentive Townvehicle, which is a human-centered and context-aware vehicle that makes people's urban lives more comfortable and enjoyable. We created several types of AT and performed some experiments on communication and coordination among several ATs using adhoc wireless networks. AT also contributes to integration of physical and informational environments and human community activities both in the real world and cyberspace.
4. Planning Research Projects
Now when it comes to projects we are planning in the near future, they should be strongly related to social issues, including education, medical services, transportation system, and general support system of joint operations. The following are the specific themes:
Distance Learning Systems
Studies on mechanisms and environments to learn anytime, anywhere, and in anyone's pace
Medical Information Systems
Studies on methods and systems to search and mine knowledge about medical matters
Urban Transportation/Information Systems
Studies on new vehicles and transportation systems in which vehicles are not only physical movement support tools but also physical-context-aware information systems
Collaborative Creation Support Systems
Studies on theories and practical systems to support human collaborative works and creation of innovative ideas
It is obvious that quite a few researchers have already engaged in these themes. This implies they are very challenging. It is common, however, that things are often simpler than they actually are if we see them from a slightly different angle. There should be some points which many people overlook. So we would like to find them out carefully.
5. Research Environment
Each student is given a notebook computer with built-in wireless LAN, a desktop computer and LCD. He/she is allowed to use a PDA with built-in wireless LAN, too. But a desktop computer is only given to a student who can build it with parts by himself/herself.
As common machines shared with all the members, several server machines embedded RAID disc is installed in this lab. The mail server is a Cobalt Cube made by Sun Microsystems and its blue box-shaped figure is pretty cute.
Three spaces are available for the students. Other than a student area and a meeting area, there is a Japanese room which has tatami mats, where we can relax like home and can make various things. Also the student area is functioning as a living space for them equipped a fridge, a microwave, and so on. It stores food and treasured plum wine (Umeshu), which we can purchase with PDA. Although the research environment is far from the ideal, we are willing to improve it little by little so that we all enjoy working.