Keynote and Invited Speakers
Prof. Dr. Akinori Nishihara
Department of Human System Science,
Graduate School of Decision Science and Technology,
The Center for Innovative Teaching and Learning (CITL),
Tokyo Institute of Technology
Joy of having double majors
Akinori Nishihara is a professor at Department of Human System Science, Graduate School of Decision Science and Technology, Tokyo Institute of Technology. He is also responsible for Department of Communications and Computer Engineering and the Center for Innovative Teaching and Learning. Akinori Nishihara received the B.E., M.E. and Dr. Eng. degrees in electronics from Tokyo Institute of Technology in 1973, 1975 and 1978, respectively. Since 1978 he has been with Tokyo Institute of Technology, where he is Professor of the Department of Human System Science, Graduate School of Decision Science and Technology. He also belongs to the Center for Research and Development of Educational Technology. His main research interests are in one- and multi-dimensional signal processing, and its applications to educational technology. He has published more than 200 technical papers in international journals and conferences. He served as an Associate Editor of the IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences from 1990 to 1994, an Associate Editor of the Transactions of IEICE Part A (in Japanese) from 1994 to 1998, and then Editor-in-Chief of that Transactions from 1998-2000. He was an Associate Editor of the IEEE Transactions on Circuits and Systems II from 1996 to 1997. He has been serving in IEEE Region 10 Executive Committee, as Student Activities Committee Chair (1996-1996), Treasurer (1999-2000), Educational Activities Committee Chair (2001-2004), and Strategic Planning Coordinator (2007-2010). He also served as an Executive Committee Member of IEEE Tokyo Section (1995-2010) and IEEE Japan Council (1999-2010). He served as a member of the Board of Governors, IEEE Circuits and Systems Society during 2004-2006.He was Chair or the IEICE Technical Group on Circuits and Systems from1997 to 1998, and served as an Advisor of that Technical Group from1998 to 2011. He served as Vice President, Service Activities of the IEICE Engineering Sciences Society, Director, Conferences, Student Activities and Education during 2007-2009, and Director, General Affairs (2011-2013). He received Best Paper Awards of the IEEE Asia Pacific Conference on Circuits and Systems in 1994 and 2000, a Best Paper Award of the IEICE in 1999, and IEEE Third Millennium Medal in 2000. He also received a 4th LSI IP Design Award in 2002, Distinguished Service Award for IEEE Student Activities in 2006, and Best Teacher Award from Tokyo Institute of Technology in 2009. Prof. Nishihara is a Fellow of IEEE, a Fellow of IEICE, and a member of APSIPA, AACE (Association for the Advancement of Computing in Education), Japan Society for Educational Technology (JSET), and Japanese Society for Information and Systems in Education (JSiSE). His research interest includes Signal Processing andEducational Technology. He has published more than 300 technical papers in refereed international journalsand conferences. He is currentlyan Institute Councilor and the Chair of TAIST Steering Committee.
Prof. Akinori Nishihara has double majors; signal processing andeducational technology. Having two majors he could broaden hisresearch more than double. With his double majors, his lab has hadconsiderable diversity in age, gender, ethnicity, and culture in itsmembers. From that environment Prof. Nishihara learned a lot.In this talk he will introduce some of his research activities, andhis life during the past 47 years in Ookayama campus of TokyoInstitute of Technology.
Prof. Dr. Hiroaki Kunieda
Dept. of Integrated Systems and Communications
School of Science and Technology
Tokyo Institute of Technology
VLSI and its Future
Dr. Hiroaki Kunieda who works in Tokyo Inst. of Technology in Japan for last 38 years since he graduated from phD course in the same university in 1978. He has engaged in education of both undergraduate and graduate school of his institute by giving lectures and experimental jobs to his students. However, his major works in his institute is to make research for the design of Integrated Circuits, which are semiconductor chips with functions for video, audio and communications. In 1995, He has worked to organize a nationwide project to establish VLSI design centers. In conjunction with this project, he also succeeded to establish a VLSI design center in our institute, which contributed a great deal the education and the research for advanced VLSI design. He has published 5 textbooks and more than 150 research papers in the academic journals or international conferences as results of his research. He worked his research with his students in my laboratory, which enabled the students to get phD degrees. More than 30 phD students graduated from his laboratory so as to work actively in academic and industrial world. In the academic society, he has engaged in many roles, especially organizing the technical conferences for both domestic and international engineers. In 1988, he has served a general chairman of Domestic conference in the field of circuit and systems in the biggest Japanese Academic Society IEICE for Electrical and Electronic Engineering. After that, he was involved in organization of the domestic conferences every year. In 1996, he was selected as a chairman of technical group of the Society IEICE to be a leader in the field of VLSI (IC) Design Automation. In 2004, he was conferred the title of "Fellow member" by IEICE Society. As my international academic activities, he was elected as a board member of the world biggest academic society IEEE Circuit and System Society in 1995. He has served a technical program chairman of many IEEE sponsored international conferences. In 2007, he joined the establishment of a new graduate school TAIST in Thailand, which shows one of his results for the international research and education support to Asian countries. Until today, he is serving a department head for TAIST ICT related field. From 1996, he started up a venture business while they remained the position of Professor in our institute. He has selected fingerprint authentication technology to identify individuals and continue to run the company until today. One of the major results is that their company succeeded to sell 1 million solutions to Sharp mobile phones in 2008. The venture is the first successful university venture in the field of Information and Communication Technology (ICT), which will become one of the best model case of the university venture in Japan.
As a memorial talk of retirement of Tokyo Institute of Technology, I would like to summarize the results of 38 years researches on VLSI design and the expected results of the future research on VLSI industry. According to the current progress of VLSI in term of size and system, the target field would be categorized to Intellectual Information system of VLSI design. A fingerprint authentication venture entrepreneurship with corporate activities of last 20 years are also summarized. The activities of 8 years for TAIST ICTES program are talked with the personal opinion about the future direction.
Asst. Prof. Akkarat Boonpoonga
Department of Electrical and Computer Engineering,
King Mongkut's University of Technology North Bangkok
Signal Processing and Its Implementation for EM Applications especially GPR and Radar Target Identification Systems.
Asst. Prof. Dr. Akkarat Boonpoonga received D.Eng.Degree from King Mongkut's Institute of Technology Ladkrabang (KMITL). His research interest includes signal processing and FPGA implementation for EM applications such as ground penetrating radar (GPR), radar target identification and smart antenna.
In electromagnetic(EM) applications such as a radar system, a signalsuch as short pulse or stepped-frequency signal is transmitted to the object. Then, the signal scattered from the object is utilized to analyze its characteristics for classification or identification.To achieve this system, the signal processing is required. In this talk, some signal processing methods which often used in EM applications will be discussed. The example of these methods are imaging subsurface of GPR system, identification technique useful for classifying object buried under the ground and calibration techniques of GPR etc. Moreover, the implementation point of view will be taken into account. The flexible and high-speed FPGA design technique will be introduced. Some simulation and measurement results will be exampled in order to guarantee the need of these signal processing.
Dr. Virach Sornlertlamvanich
Sirindhorn International Institute of Technology, Thammasat University
Thai Text Processing and Its Applications
Dr. Virach Sornlertlamvanich received his bachelor and master degrees in engineering from Kyoto University in 1984 and 1986 respectively, and doctoral degree in Computer Science from Tokyo Institute of Technology in 1998. He was esteemed "The Researcher of the Year 2001" by the Nation Newspaper (Thailand), and was conclusively awarded by the National Research Council of Thailand as "The Most Outstanding Researcher of the Year" in Information Technology and Communication in 2003. His research interests include Natural Language Processing, Machine Translation, Data Mining, Artificial Intelligence. His recent efforts are on the research of the technologies for digital content creation and social media understanding.
Talking about the language, which has no explicit word and sentence delimiter, is extremely hard for processing on computer. The text flows as a stream of conversation with arbitrary pause of the peaker. The Thai language is one of the good examples of the languages, which need preprocessing of word and sentence segmentation in the early step of morphological analysis. The current research has reported that the accuracy of word segmentation can be achieved at the rate of higher 98% depending on the text set. However, it is still behind the accuracy of human judgement. By the way, beyond the problem of word segmentation, the needs of text processing in higher level become more significant in today's growth of data, especially in terms of text data, in the cyberspace. The talk will be formalized into three fundamental problems in Thai text processing. Those are word segmentation, named entity recognition or keyword extraction, and semantic relation extraction. In the flood of information today, we spend most of the time to grasp the essence of the information rather than to enjoy the reading. Many approaches have been proposed to handle these fundamental issues, however, there is still much room for improvement. The introduced approach is not the best one, but it is aimed to make the problem well recognized. Mutual information and entropy are effective measures to uncover the possible word boundary for the non-segmenting languages such as the Thai language. It is remarkably to note that with the approach, the result has shown that about 30% of the extracted words are not defined in the Thai- Thai dictionary published by Thai Royal Institute in 1982. Keyword labeling is also a task that we can effectively apply a machine learning approach such as MIRA (Margin Infused Relaxed Algorithm) to capture the word context. This can be done on the result from the word segmentation task. Undoubtedly, the accuracy of the annotated tag is ranked from person (PER), date (DAT), location (LOC), and organization (ORG). This is because tag for person has the least ambiguity. The pattern for extracting the semantic relation between the type-annotated keywords is accordingly assigned to the word form of the disambiguated verb phrase. The experimental result shows that most of the distance between the keyword and the target verb phrase is not more than one word. Therefore, we can find the target verb phrase in the adjacent position or one word skipped position with the highest probability. Based on the solution for the above NLP fundamental issues, many more tasks are made possible on the current viable Internet connection. The talk demonstrates the three constructive applications on the huge generated data i.e. linked data formation for knowledge map reasoning; keyword tracking on social media to understand the online social movement; and hyper local news publishing to fill in the information gap between urban and rural life. The task of natural language processing today is not just only for the language itself any more, but it can bring along the possibilities on the advance of the Internet, big data, and machine learning technique.
Prof. Dr. Manabu OMAE
Graduate School of Media and Governance, Keio University
5322, Endo, Fujisawa, Kanagawa, 252-0882, Japan
Towards practical use of automatic driving vehicles - Issues to be considered and solved from a researcher's point of view
Manabu OMAE is a professor of the Graduate School of Media and Governance, Keio University in Japan. He was born in 1972, and received his bachelor's degree, master's degree and Ph. D. in Mechanical Engineering from the University of Tokyo in 1995, 1997 and 2000. He has worked for Keio University since 2000. His research field is automotive engineering, especially Intelligent Transport Systems (ITS), vehicle control for automatic driving and platooning.
Interest in the development of automatic driving vehicles that improve the safety, efficiency, and accessibility of the transit services and quality of highway travel is greatly increasing. It is expected that a passenger vehicle with automatic driving function on the expressway will come onto the market about 2020. In the lecture, some of our research works on automatic driving systems are introduced and issues to be considered and solved towards practical use and spread of automatic driving vehicles in the society are explained from a researcher's viewpoint.
Assoc. Prof. Dr. Hirohiko Kaneko
Dept. of Information Processing, Tokyo Institute of Technology
Estimation of attentional location and state based on the measurement of unconscious eye movements.
Hirohiko KANEKO is an associate professor of the Department of Information Processing, Tokyo Institute of Technology in Japan. He received his bachelor's degree in Science (Applied physics) in 1987 and Ph. D. in Engineering (Information processing) in 1992 from Tokyo Institute of Technology. His research field is vision sciencerelated to space perception, multi-sensory interaction and eye movements, and its applications.
We found that characteristics of unconscious eye movementswere modulated by attentional state of observer. Pupillary response to stimulus luminance wasmodulated by the strengthof attention to the stimulus. Small vergence eye movement wasinduced by paying attention to an object moving in depth while fixating a stationary object. Optokinetic nystagmus (OKN) occurred by paying attention to an area with uniform motion. The magnitudes of the eye responses in these cases were smaller than those in the case of direct viewing, but the direction of responses were the same as those in the direct viewing. From these facts, it is possible to estimate the location and state of visual attention from the state of the eye movements while viewing a visual scene containing objects or areas with various luminance, depth and motion. In this talk, I will show the details of the relationship between attentional state and eye movements, and mention the system to estimate the location and state of visual attention based on the measurement of unconscious eye movements.
Asst. Prof. Dr. Ikuhisa Mitsugami
8-1 Mihogaoka, Ibaraki, Osaka, 567-0047, JAPAN
Human Motion Analysis for Intention-Gait Modeling
Dr. Mitsugami received the B.S. degree in Engineering from Kyoto University in 2001,and the M.S. and Ph.D. degrees in Engineering from Nara Institute ofScience and Technology in 2003 and 2007, respectively. He is currentlyan Assistant Professor with the Institute of Scientific and IndustrialResearch, Osaka University. His research interests include geometry incomputer vision, detection and tracking, gait analysis. He is a memberof the IEEE, IEICE, IPSJ, RSJ, and VRSJ.
Our research group have studied human gait (way of walking) for more than ten years. Considering the fact that gait is unique for each person, we have regarded it as one of the biometrics and applied it to person authentication task. Fortunately, our achievements get a great attention from the society and are getting to be used in real crime investigation. Equipped with the achievements and experiences, we launched a novel research project "Behavior Understanding based on Intention-Gait Model" supported by JST-CREST in 2010. In this project, we focus on a new aspect of gait that the gait of a person is not constant; he/she often changes his/her gait affected by emotion, physical/mental conditions, and relationship among people around him/her. We regard such inertial information as his/her "intention," and try to solve relation between the intention and gait. In this talk, I briefly overview this project and describe details of some topics such as 3-D modeling of a walking person, depth-based gait authentication, attention estimation, and impairment detection.