Takashi Tomii

Design of video scene databases with mapping to virtual CG space

This paper presents a new approach of modeling video scenes by adding meaning to the shape of real world objects and simplifying it. In order to store shapes and/or motions of objects that appear in real world scenes into the database, expressing the shape and its meaning and handling them efficiently are indispensable. For that purpose, we introduce a method for constructing a scene database based on an approximate polygon model that reflects the individual shape and meanings of an object, called shape of individual model. It is particularly suitable for modeling the human head. The scene database is constructed using the Shapes of Individual Models and their 3-dimensional motion by mapping then into virtual CG space. And this paper also discusses the evaluation of the created Shape of individual Model in an application of the scene database.

Motion Simulation of the Human Workers for the Integrated Computer-Aided Manufacturing Process Simulation Based on Info-Ergonomics Concept

This paper presents modeling of Working Process and Working Simulation in factory works. Also a concept for Mediator-based human body/motion modelin for application to Info-Ergomomics is offered. In Manufacturing Process Design, many simulations are executed in order to improve efficiency of designer’s work. But in many cases they are focusing on simulating and evaluating machines’ performance. In the present paper we will propose Info-Ergonomics, and Working Simulation by using CG. Also, we consider it as a possible way for unifying all the data used in various applications (CAD/CAM, etc) during the design process and evaluating all subsystems in a virtual Factory.

Human-Body Motion Simulation Using Bone-Based Human Model and Construction of Motion Database

This paper presents motion simulation/evaluation system for factory workers in the framework of “Info-Ergonomics.” One of the key technologies is CG simulation based on the precise human body mockup called “Bone-Based Human Model.” Using BBHM, “real” motions of workers can be mapped for precise simulation. Another important issue is data and knowledge integration. For the purpose of schematizing such data and providing retrieval functions are discuss in an extended database system, “Real World Database.”

Data Analysis Framework for Visualizing Correlation of Energy Consumption and Transit Time in Road Sections using the ECOLOG database

We have built a database to estimate electric vehicle (EV) energy consumption from driving logs. This study, examines a data analysis framework that aggregates numerous operation logs that are grouped by predefined road sections which expert drivers consider to have meaning. It visualizes correlation of energy consumption, transit time, and specific factors of energy loss. Furthermore, we estimate energy loss reductions by improved driving when drivers obtain feedback from data analysis using our proposed frame.

Accuracy Evaluation by GPS Data Correction for the EV Energy Consumption Database

Electric vehicles (EVs) are expected to be applicable to smart grids because they have large-capacity batteries. It is important that smart grid users be able to estimate surplus battery energy and/or surplus capacity in advance of deploying EVs. We constructed a database, the Energy COnsumption LOG (ECOLOG) Database System, to store vehicle daily logs acquired by smartphones placed in vehicles. The electrical energy consumption is estimated from GPS coordinate data using an EV energy-consumption model. This research specifically examines commuting with a vehicle used for same route every day. We corrected GPS coordinate data by map matching, and input the data to the EV energy consumption model. We regard the remaining battery capacity data acquired by the EV CAN as correct data. Then we evaluate the accuracy of driving energy consumption logs as estimated using the corrected GPS coordinate data.

Introduction and Implementation of a System to Share Findings with ROI for PET Diagnosis Support

Positron Emission Tomography (PET) is an effective imaging technique to detect cancers, but interpretation of PET images is difficult for radiologists who have insufficient knowledge and experience. Radiologists could interpret PET images more effectively if they were able to share their PET expertise. To realize this goal, our recent efforts have designed a PET information model that represents the process of interpreting PET images; we have implemented the Whole-Body PET Database (WB-PET DB). The PET information model defines the WB-PET DB schema. WB-PET DB stores findings along with a region of interest (ROI), thereby allowing ROI image retrieval by keys that radiologists use when writing findings. Regarding WB-PET DB, we have shown its conceptual model in recent works, but have revealed little about its system architecture. In this study, we designed and implemented the WB-PET DB System. This system contains the user interface to insert ROI into WB-PET DB and the component that outputs results of retrieval. This component outputs them in a DICOM file format; thereby users can view stored ROIs using 3D visualization software such as AmiraTM. Results of this study demonstrated a model to retrieve an image of body functionality as a PET image.

Extracting Event Semantics from Video Data Based on Real World Database

Video is considered the most effective medium for capturing events. There are two major approaches of representing video contents: the structured modeling approach and stratification. They require video to be divided into simple semantic units on top of which some structures are built to express high-level video semantics. Text is used to describe the necessary semantic units. A major problem of such an approach is that the descriptions tend to be incomplete and subjective. In addition, segmenting video sequences and describing their semantic contents is a tedious task.