The 6th COE Postdoctoral and Doctoral Researchers
Technical Presentation

Program (20 mins each: 15 mins presentation and 5 mins discussion)

1. "Nara Palace Site Navigator: A General Human Navigation System Using a Networked Shared Database"
   天目 隆平 (視覚情報メディア講座: D2)
   Ryuhei Tenmoku (Vision and Media Computing Lab.: D2)
   

   [Abstract]
   Many human navigation systems which present location-based information to mobile device users are developed in recent years. We have already developed a wearable tour guide system which presents location-based tour guide information using augmented reality techniques. In this research, we try to improve the previous system into a general human navigation system "Nara Palace Site Navigator" which can present tour guide information according to the specifications of the user's devices: wearable computer, personal digital assistant (PDA), and cellular phone. Additionally, this presentation describes the experiments using the proposed system in the Nara palace site to show the feasibility of the proposed system.
   
   Top

2. "Remodeling of the Sensor for Non-Audible Murmur (NAM) Interface Communication"
   中島 淑貴 (言語科学講座: D2)
   Yoshitaka Nakajima (Applied Linguistics Lab.: D2)
   

   [Abstract]
   We are working towards the next generation of skin-attachment sensors for sampling NAM signals (Non-Audible Murmur) by using soft silicone, which has an acoustic impedance close to that of human flesh, as the prime medium of vibration. These new NAM microphones enable us to record more wideband NAM sound, which is clear enough to convey articulated utterances even without digital signal processing, and provides increased robustness against noise in comparison with the former stethoscopic type. We obtain a much higher accuracy of NAM recognition and, as a result, suggest the possibility of a "Non-Voice Phone".
   
   Top

3. "Blind Decomposition of Binaural Mixed Signals Using Multistage SIMO-ICA"
   高谷 智哉 (音情報処理学講座: D1)
   Tomoya Takatani (Speech and Acoustics Lab.: D1)
   

   [Abstract]
   We address the blind decomposition problem of the binaural mixed audio signals to realize the audio augmented reality system. In order to solve the problem, we have proposed Single-Input Multiple-Output (SIMO)-model-based indepednet component analysis (ICA) (SIMO-ICA) algorithm which decompose the mixed observed signals into the SIMO-model-based signals. Here the term "SIMO" represents the specific transmission system in which the input is a single source signal and the outputs are its transmitted signals observed at multiple microphones.

   In this presentation, in order to improve the decomposition performance, we propose the learning algorithm combining frequency-domain ICA with projection back processing and SIMO-ICA. To evaluate its effectiveness, decomposition experiments are carried out under a reverberant condition for the different directions of sources. The experimental results reveal that the decomposition performance of the proposed method is superior to that of the conventional method.

   
   
   Top

4. "Driver's reliance on a car crash prevention system"
   黒崎 章 (システム制御・管理講座: D1)
   Akira Kurosaki (Systems and Control Lab.: D1)
   

   [Abstract]
   A driver assistance system may have bad effects on drivers if they may rely totally on the system. On the other hand, drivers may refuse to use the system if they feel it unnecessary. In this study, we conduct experiments to estimate the degree of a subject's dependence on a driver assistance system. Subjects operate a driving simulator equipped with a system that warns them if an obstruction exists ahead. In these experiments, they experience various patterns of system behavior, including warning malfunction while driving. We collect and analyze the driving simulator's behaviors while driving to estimate the degree of the subjects' dependence on the system.
   
   Top