Mobile Stereoscopy
Project Members:
Yuta Nakayama, Steven Zhou
Idea:
Mobile Stereoscopy project aims to develop a networked 3d stereoscopic content creation and sharing environment. As a part of the Human-Computer Interaction research, stereoscopy adds a modality of depth and spatial feeling to interactive systems. Though we live in three-dimensional space, our current dominant media systems such as pictures, TV and also internet web pages, are evolved in two-dimensional forms. Our final goal is to construct a low-cost and easy-to-use immersive communication environment, such as 3d version of YouTube or Flickr.
The goals of our research are to build:
• Mobile stereoscopic image/video acquisition device
• A web framework for viewing and sharing stereoscopic contents.
• Algorithm for creating and displaying stereoscopic contents optimized for mobile web application. |
|
Field Analysis of Open Source OCR Engine Using Mobile Devices
Project Members:
Syed Omer Gilani, Steven Zhou
Idea:
Mobile phones have evolved from passive one-to-one communication device to powerful handheld computing device. Today most new mobile phones are capable of capturing images, recording video, and browsing internet and do much more. Exciting new social applications are emerging on mobile landscape, like, business card readers, sing detectors and translators. These applications help people quickly gather the information in digital format and interpret them without the need of carrying laptops or tablet PCs. However with all these advancements we find very few open source software available for mobile phones. For instance currently there are many open source OCR engines for desktop platform but, to our knowledge, none are available on mobile platform. Keeping this in perspective we propose a complete text detection and recognition system with speech synthesis ability, using existing desktop technology. In this work we developed a complete OCR framework with subsystems from open source desktop community. This includes a popular open source OCR engine named Tesseract for text detection & recognition and Flite speech synthesis module, for adding text-to-speech ability.
|
I-Explore
Project Members:
Lor ZhiChang, Miriam Schneider, Jayashree Karlekar, Daniel HII Jun Chung, Steven Zhou
Idea:
Abstract: Exploring buildings, streets or entire cities from the past is currently only possible in a museum, and is often presented in the form of static displays (models, plans, drawings, photos). Similarly, designs of new buildings are generally presented away from the actual building site, and abstract plans or at best models are used. This stretches the viewers' imagination, as it is often difficult for them to relate these displays to the current location and looks, and to immerse themselves in the scenes.
The iExplore project aims to alleviate these problems by developing a mobile hand-held system that can be used to view buildings, streets, historic sites, landscapes or cityscapes on location, and for interactively exploring them through time (going into the past and the future) as well as space (by the user walking around the city, for example). The look of the objects in the past or future would be overlaid onto the present objects in the mobile device's display.
Objectives:
Develop an accurate, mobile location detection and movement tracking technology for the mobile device that work on a coarse scale (where is the user) as well as a fine scale (where is the user looking). Develop an accurate registration method for aligning virtual and real objects in the mobile device. Design architectural models suitable for exploration. Develop an intelligent client-server architecture that helps in location detection and allows users to share location-specific photos and videos from different time periods as well as models of future developments.
Daniel Hii's role in the iExplore project is to provide his expertise and knowledge in architectural and urban design as well as their detailing. The architectural content for the project from 2D mapping to 3D modeling, site 2D information and the textures mapping required are all part and parcel of his work. He is also involved in the testing of linking all these data for the usage of mobile 3D model visualization. Explorations are required to get the most efficient 3D modeling and texture format to be run on mobile devices as they have less processing power, memory and storage capacity than their desktop counterparts. From there onwards, other possible features can be explored to enhance realism which includes adding shaders and particle systems for the 3D graphics generated.The core of his job is to make sure the represented 3D models are accurate and correct architecturally.
The applications designed to test our new motion-based interfaces are explained below:
|
Car Racing:
To steer a car in the game, the user turns the phone like a steering wheel. Moving it back and forth controls the speed. |
|
Map Navigation:
The phone acts as a magnifying glass over a virtual map. Moving the phone forwards and backwards zooms through different resolutions of the map, and moving the phone along the map plane scrolls it in any direction. |
 |
|
|
Last Updated on Wednesday, 18 November 2009 20:54 |
|
|
|
|
|
