Demo section

The Demo section of the 16th FRUCT conference will be combined with the conference social event. It will be held on October 30 (Thursday) from 17.30 till 20.30. The conference and seminar participants are warmly welcome to take part in the event.

We ask all authors to send titles and 2-3 paragraph descriptions of your demo proposals to infofruct.org by October 6, 2014. Please also make sure that the demo will be prepared in line with the below provided recommendations.

All authors planning to present projects at FRUCT demo session must submit the presentation slides (in PDF), plus name and affiliation of the presenter to emails ilya.paramonovfruct.org and infofruct.org; with Subject: Poster presentation #xx (demo ID in the list below) by October 27, 2014.

Introduction

The demo section at FRUCT conference will consist of two parts. The first part is the promotional section, which is used to present/introduce demo projects to the public. The main idea of this section is to make people aware of the main point of the demo and be interested enough to come to demonstration stands during the second part of the demo session.

The second part of demo session will be held in classical stand-based format, where the team will get a place to install the demo and place for a poster. If you have some special requirements please contact organizing committee by email infofruct.org.

The demo section of FRUCT conference will be held in Pecha Kucha format of presentation. This page contains all required information about the Pecha Kucha format and requirements that have to meet.

Pecha Kucha Presentation Format

Pecha Kucha is a presentation technique where a speaker shows a definite number of slides, each for 20 seconds. The slides are changed automatically during the talk. The main intention for Pecha Kucha presentation style is to prevent participants from being too verbose and to make their talks more dynamic and impressive.

Pecha Kucha Night is an event where each speaker uses Pecha Kucha presentation, and speakers change each other in non-stop fashion. Initially invented by architects, this kind of event is often used to present creative projects or work; nowadays it is also used for R&D talks too. Pecha Kucha Night format allows all participants to make announcements about their demos in attractive and time-efficient way. That is why we have chosen this format for demo promotion section at FRUCT conference.

How to prepare Pecha Kucha presentation

Here is an instruction on how to prepare your Pecha Kucha style presentation for Demo promotion section. The essentials points of this instruction are also available in the form of check list below.

Your presentation must contain exactly 6 slides, and each of them will be displayed for 20 seconds. The slides will be changed automatically. So, the whole presentation will take exactly 2 minutes (it should be noted that usually Pecha Kucha presentation has 20 slides, but we have to reduce number of slides due to a large amount of submitted presentations).

Provide the information about yourself and your presentation on the first slide (name, institution, title of your presentation).

The main purpose of your talk would be to interest people, so your presentation should make absolutely clear the main ideas of your project and explain what you plan to show at the demo stand. Make your presentation fascinating to attract attendees and avoid technical details in your talk.

Reveal one main idea on each slide. Do not overload your slides with information. Remember, that each slide is displayed only for 20 seconds. Place no more than 2 lines of text per slide, or one big picture. Do not use slide titles.

Do not duplicate the same slides in your presentation — it is cheating! If you see that 20 seconds for a particular slide is not enough for you, try to decouple it into the two or more, or omit the details.

Do not place “Thank you” or “Q&A” slides in the presentation. Pecha Kucha session does not imply any questions from the auditory. All the questions will be asked afterwards in a poster room.

Prepare your speech thoroughly and beforehand. As you have only 20 seconds per slide, it is quite impossible to improvise during the talk.

Rehearse your speech several times to be sure in the absence of pauses when you wait for the slide change, or accelerations when you fails to follow your slides. Try to speak in the same pace during all the presentation. It definitely depends on your text, so try to prepare near the same amount of text in speech for each slide.

Check list

Use exactly 6 slides.
Place information about yourself and your presentation (name, institution) on the first slide.
Reveal one main idea on each slide.
Place no more than 2 lines of text or 1 large image per slide.
Do not duplicate the same slides, do not place “Thank you” or “Q&A” slides in the presentation
Do not use any slide change animation
Prepare your speech thoroughly and do not forget to rehearse it

Useful links

LIST OF ACCEPTED DEMOs

Training Assistant: the Automatic Training Data Collection System, by Ivan Timofeev, Sergey Lobarev, Nikita Timofeev and Sergey Averkiyev, YarSU, Russia
Intelligent Mobile Tourist Guide: Tourist Assistant - TAIS, by Maxim Shchekotov, Nikolay Teslya, Alexei Kashevnik and Andrew Ponomarev, SPIIRAS, Russia
Smart Space-Based Lego® Mindstorms EV3 Robots Interaction, by Nikolay Teslya, SPIIRAS, Russia
Kizhi Island virtual model, by Matti Pouke, University of Oulu, CIE, Finland
Control Your Things: Tool for Collection and Control of Personal Things, Ksenia Lagutina, Andrey Vasilev and Ivan Timofeev, YarSU, Russia
Accessibility map mobile application, Irina Shabalina and Kirill Kulakov, PetrSU, Russia
Monitoring system for patients with hypertension based on mobile application and cloud service, Ivan Shchitov, YarSU, Russia
System for monitoring patients with diabetes, Maxim Yatskovskiy, FRUCT MD LLC, Russia
A system for automatic verification of arrhythmia detection algorithms by Artem Rudenya, Alexander Borodin and Yulia Zavyalova, PetrSU, Russia

Description of Accepted demos

A lot of people visit gyms to compensate their insufficient activity level. They record completed exercises to diaries during workouts. This data can be used as a basis for development of individual training program that allows to increase productivity of next workouts. However, the manual addition of exercise records to the diary takes a lot of time and may be error-prone due to person’s fatigue. At the demo section we demonstrate a part of the use case we show how server collects exercise data and composes it into the exercise record. We developed the prototype of the device on gym apparatus, the model of an exercise machine and the prototype of the server application that demonstrates data processing.
Tourist Assistant - TAIS (Tourist Attraction Information Service) is a mobile tourist guide developed for Android-based devices. It provides tourists with recommendations on what attraction (e.g., museum, monument, social place) are currently better to visit, based on the tourist’s preferences and current situation in the region (e.g., closed attractions, weather situation, accessible drivers around the tourist who can drive him/her to places of interest). The service is based on Smart-M3 information sharing platform that provides the smart space infrastructure. Recommendation methods, algorithms, and scenarios have been successfully developed, prototyped. Tourist Assistant - TAIS provides possibilities for tourists browsing attractions descriptions and photos, estimating attractions and related images to improve recommendation performance. For information sources the following resources are used at the moment: Wikipedia, Wikivoyage, Panoramio. Moreover, the service can display current user location on the map, provide possibilities to build pedestrian and car paths to the interested attraction, find fellow travelers who can pick tourist up around location and drop off around interested attraction. Current weather in the tourist location is displayed for the tourist in mobile device and used for making recommendations for the tourist (e.g. in rainy weather outdoor attractions less preferred than indoor). Mobile tourist guide consists of several services that solving particular tasks and interact in smart space that allows to provide interoperability support between them. There are attraction information service, recommendation service, administration service, context service and mobile client. Attraction information service extracts the information about attractions from different Internet sources and shares it with the smart space. Recommendation service takes the lists of shared attractions, their images and descriptions and range this lists according the tourist preferences and context situation in the considered area shared with the smart space by context service. It uses for automated filtering and ranking two approaches: content analysis and user evaluation. Administration service allows to setup the following parameters for the mobile tourist guide operation: attractions searching radius, recommended attractions count, smart space waiting time, new default items count, images searching radius for main page, GPS inquire timeout, and changing location for GPS inquire. Using the mobile client for accessing to the mobile tourist guide allows the tourist to see recommendation about attractions in the region during the trip and rate the attractions, photos and their descriptions. In the Fig. 1 the architecture of mobile tourist guide is presented. Mobile client shares with the smart space the tourist context information like coordinates and his/her preferences. Attraction Information Service acquires from different Internet sources attraction information (a list of attractions around the tourist, images and their descriptions). The recommendation service analyzes the attraction information and tourist preferences, estimations and provides recommendations for the tourist. Mobile client shows these recommendations for the tourist. Administration Service allows to set special settings, which allows to tune mobile tourist guide with balance between quickness and recommendations quality. Context Service provides weather information in the region around the tourist.
Nowadays cyber-physical environments are spreading wide all over the world. There are many systems such as “smart home”, “smart car”, “smart city”, which are based on the real time interaction between physical world and cyber world. One of the examples of cyber-physical system is a “smart home” devices interaction for home cleaning. For example, there are two robotic vacuum cleaner in the house. Before the cleaning they creating a map of dirtying and split the house to the two parts based on the nearest dirty points. The other example is a situation when vacuum cleaner finds a hindrance, for example big toy or chair. It shares information about the situation through the smart space and manipulating robots should decide which of them would go to remove the hindrance. The demo provides interaction model of devices in smart space. These devices are robot cars based on the Lego® Mindstorms EV3 constructing kit with ultrasonic and gyroscope sensors. Each car is driven by two independent large motors and controlled by control block with LeJOS installed on the SD-card. WiFi USB-adapter provides local area network connection. For the control block the LeJOS has been chosen because it is provide full functionality OS with JRE Environment. Robots shares information through the Smart-M3-based smart space via using the Java KPICore library. After the start command receiving from the smart space, each car rotate at 360 degrees and fetch information about turn angle and distance from the gyroscopic and ultrasonic sensors correspondingly. This information stores for the future processing to separate found objects. The separation algorithm discerns objects by analyzing difference between neighbor distances. All found objects are sorted by the average distance and shared through the smart space. Each robot has subscription to new objects that are appearing in the smart space. When the objects from the other robot appear, they are queried by listening robot and comparing with existing set of its objects. The values equality means that robots find each other and these values should be excluded from the set. If there are more than two equalities, robots should move toward one of the equal objects and repeat the area scanning. When robots find each other, they can detect the object and select the nearest robot. This robot turns to the angle the object is located in and moves to the object. Additionally the demo provides an Android-Based application for robot control through the smart space. This application allows starting and ending robot moving, changing direction of moving (forward or backward) and turning around for object searching as well as showing the state of each robot. Robot shares each action throug the smart space. These actions are: moving forward or backward, turning to the right or to the left, scanning area, objects discerning, information sharing and making a decision. Two nearest objects are showing ih the corresponding fields.
The Kizhi 3D virtual model is a multi user virtual world depicting the southern section of Kizhi Island at lake Onega, Russia. The model is hosted on Meshmoon service.It was modeled using the free and open 3D creation software Blender. Principles of low-polygon modeling were used to ensure the model’s applicability to real-time interactive 3D virtual environments such as the RealXtend Tundra. With low-poly modeling, the complexity of the model is kept to minimum which ensures viewing the model in interactive applications is not too computationally intensive. The main focus of the modeling was the Pogost area containing five major buildings which were The Church of the Transfiguration of Our Savior, The Church of Intercession, The Bell Tower of the Kizhi Pogost, The Fence of the Kizhi Pogost as well as the house of peasant Oshevnev. Main detail of the modeling work was focused in the Pogost area and its immediate surroundings. In its entirety, the model ranges from the hydrofoil docks in the north to the southern end of the island.
Everyone faces the challenge of finding and collecting all things for the travel, journey, etc, and tracking them to ensure that they are not forgotten or lost. One possible solution for these tasks is to attach electronic tags to required things and to keep a track of them by a smartphone or a special scanner. There are 4 technologies that can be used to create electronic tags: Wi-Fi, Bluetooth, RFID, NFC. All tags differ from each other in the size, range, battery life, format of the recorded data and other technical characteristics. These parameters define how the user’s smartphone can interact with electronic tags and which use cases can be implemented in the application. For example, the data exchange between RFID-tags and smartphone requires a scanner, but other types of tags can be directly used by the phone. The proposed system for control of personal things consists of a mobile application and electronic tags. The user attaches the tags to things. The database of things with tags is stored in the phone. Also if the user wants to keep a track of objects with RFID-tags, one will be able to use the special scanner. It can be put in the bag, drawer or any other container where user wants to track things. We have designed three use cases for the system: 1) The user collects things in a bag. The application automatically creates a list of detected things. Then one can rename, delete, edit list contents and check presence of things later on.; 2) The user chooses the thing that one cannot find. The application determines and shows to the user the distance between the smartphone and the tag associated with the thing. When user is on the move the displayed distance is changing accordingly; 3) The application can be set to alarm the user if the thing with the tag disappears from the visibility range. Currently the application implements list creation and things presence checking use cases. In the first scenario the user sees on the screen the names of all objects which have been added previously for tracking. Then one manually forms the list by selecting thing’s names. In the second scenario the user uses a list that was created in advance. When one wants to collect things into the bag he/she chooses the appropriate list and carries things with attached tags near the phone. The application highlights the corresponding list items and puts them to the end. When all things were detected the user sees the message “All things are collected”. Current version of Control Your Things application supports electronic tags with NFC technology. In the future authors plan to add support for Bluetooth and RFID tags and implement other use cases.
“Accessibility Map" is a mobile service for cartographic visualization of information about social facilities categorized by accessibility levels and types of disability. Development of “Accessibility map” is a part of project related to development of information environment for persons with disabilities. The aim of the project is to improve quality of life of persons with disabilities by addressing issues related to social exclusion, accessibility and mobility of disabled people by means of advanced ICT. Mobile application “Accessibility map” is targeted to cartographic visualization of accessibility information about social significant objects. The application utilizes data from “Accessibility Passports” web-service, which database is filled by regional social authorities. All objects of application “Accessibility map” are categorized by accessibility levels for various types of disability. The application also provides following information: name of the object or organization, description of organization activity, description of routes to an object, accessibility information related to the types of disability. All organizations are grouped by activity. The categories are healthcare, education, social care, sport, culture etc. Two means of visualization are available in the service: a categorized list of objects, and a map with geo-tags. Accessibility level of an object on the map is marked by color. Also search and filtering features have been implemented in application, types of disability are marked with appropriate signs. The objects can be filtered by the several criteria: organization activity; categories of persons with disabilities (moving on a wheelchair”, “hearing-impaired”, “visually impaired”, etc.); levels of accessibility (“fully accessible”, “conditionally accessible” and “inaccessible”). The development was carried out with support of grant KA432 “Journey planner service for disabled people (Social Navigator)” of Karelia ENPI program.
Monitoring patients suffering from hypertension is a very important task that requires doctor to watch the patient’s health status regularly. Such a task consumes a lot of resources. In order to reduce the number of patient’s visits to a hospital we propose a system that allows monitoring patient’s health status remotely. The system consists of Blood Pressure Diary application, which is installed on the patient’s Android smartphone, and a cloud server. The patient can add blood pressure measurements into the diary manually or automatically via NFC or Bluetooth. These measurements are automatically synchronized with cloud that allows to systematically send information about health state to the doctor via e-mail and report to the hospital in case of emergency. Another functions of the system include keeping the same data on all of one’s mobile devices and browsing the patient’s measurements with additional context information (e.g., weather) via web UI.
The demonstration will show a system based on Cubieboard2 hardware as a home gateway mHealth, which allows receiving data from the Bluetooth glucometer and display reminders for patient on TV. This gateway also interacts with the cloud mHealth platform, which sends the measurement data and where we can to initiate a reminder to the patient.
A number of arrhythmia detection algorithms are being developed within a CardiaCare project. These algorithms should be thoroughly verified using real electrocardiograms with abnormalities annotated by experts. To provide a way of semi-automatic continuous testing of arrhythmia detection algorithms, the proposed system have been developed. Annotated electrocardiogram recordings are fetched from open databases. Short fragments are stored according to arrhythmia type identified by expert. During tests the system runs the algorithm implementation on automatically constructed set of signals and reports obtained performance evaluation indicators (sensitivity, specificity and accuracy).