Junjun Yin

Touch2Query enabled mobile devices: a case study using openstreetmap and iphone

This paper describes our mobile spatial interaction (MSI) prototype Touch2Query which presents the idea of using the touch screen on mobile devices to assist in performing ad-hoc spatial queries. This approach differs from conventional mobile LBS applications where the query shape (search space) is limited to either a bounding box or radius. Instead, we provide functionality that allows users to interactively draw any desired query shape overlaid on an area of interest directly on a mobile device with their finger by combining vector primitives such as circles, polygons, polylines, and points. With the help of location and orientation aware mobile devices, mobile maps, and real-time distance and area measurements, Touch2Query gives the users freedom to perform customised spatial queries on objects/areas of interest while realising a better contextual understanding of their spatial environment at the same time.

Mobile Visibility Querying for LBS

This article describes research carried out in the area of mobile spatial interaction (MSI) and the development of a 3D mobile version of a 2D web-based directional query processor. The TellMe application integrates location (from GPS, GSM, WiFi) and orientation (from magnetometer/accelerometer) sensor technologies into an enhanced spatial query processing module capable of exploiting a mobile devices position and orientation for querying real-world spatial datasets. This article outlines our technique for combining these technologies and the architecture needed to deploy them on a sensor enabled smartphone (i.e. Nokia Navigator 6210). With all these sensor technologies now available on off-the-shelf devices, it is possible to employ a mobile query system that can work effectively in any environment using location and orientation as primary parameters for directional queries. Novel approaches for determining a users visible query space in three dimensions based on their line-of-sight (ego-visibility) are investigated to provide for “hidden query removal” functionality. This article presents demonstrable results of a mobile application that is location, direction, and orientation aware, and that retrieves database objects and attributes (e.g. buildings, points-of-interest, etc.) by simply pointing, or “looking”, at them with a mobile phone.

EgoViz --- A Mobile Based Spatial Interaction System

This paper describes research carried out in the area of mobile spatial interaction and the development of a mobile (i.e. on-device) version of a simulated web-based 2D directional query processor. The TellMe application integrates location (from GPS, GSM, WiFi) and orientation (from digital compass/tilt sensors) sensing technologies into an enhanced spatial query processing module capable of exploiting a mobile devices position and orientation for querying real-world 3D spatial datasets. This paper outlines the technique used to combine these technologies and the architecture needed to deploy them on a sensor enabled smartphone (i.e. Nokia 6210 Navigator). With all these sensor technologies now available on one device, it is possible to employ a personal query system that can work effectively in any environment using location and orientation as primary parameters for directional queries. In doing so, novel approaches for determining a users query space in 3 dimensions based on line-of-sight and 3D visibility (ego-visibility) are also investigated. The result is a mobile application that is location, direction and orientation aware and using these data is able to identify objects (e.g. buildings, points-of-interest, etc.) by pointing at them or when they are in a specified field-of-view.

Preparing Detailed 3D Building Models for Google Earth Integration

Today’s spatially aware users are becoming more interested in retrieving personalised and task relevant information, requiring detailed 3D city models linked to non-spatial attribute data. However, current implementations of 3D city models are typically LoD2 that don’t include geometric or attribute details about many visible features (e.g. rooms) of a building. As such, value-added applications developed for web-based and wireless platforms are limited to querying for available non-spatial business data at the building level only. To overcome this, geometrically accurate 3D building models are necessary to enable users to visualize, interact, and query for task specific non-spatial business data. This paper proposes a workflow for creating detailed 3D building models with LoD3 from TLS data and uploading these models into Google Earth so that users can then explore the non-spatial business data of a building and its sub-components (e.g. windows, doors, rooms). Processing bottlenecks of the proposed workflow for detailed 3D building reconstruction are also discussed.

Spatial Search Techniques for Mobile 3D Queries in Sensor Web Environments

Developing mobile geo-information systems for sensor web applications involves technologies that can access linked geographical and semantically related Internet information. Additionally, in tomorrow’s Web 4.0 world, it is envisioned that trillions of inexpensive micro-sensors placed throughout the environment will also become available for discovery based on their unique geo-referenced IP address. Exploring these enormous volumes of disparate heterogeneous data on today’s location and orientation aware smartphones requires context-aware smart applications and services that can deal with “information overload”. 3DQ (Three Dimensional Query) is our novel mobile spatial interaction (MSI) prototype that acts as a next-generation base for human interaction within such geospatial sensor web environments/urban landscapes. It filters information using “Hidden Query Removal” functionality that intelligently refines the search space by calculating the geometry of a three dimensional visibility shape (Vista space) at a user’s current location. This 3D shape then becomes the query “window” in a spatial database for retrieving information on only those objects visible within a user’s actual 3D field-of-view. 3DQ reduces information overload and serves to heighten situation awareness on constrained commercial off-the-shelf devices by providing visibility space searching as a mobile web service. The effects of variations in mobile spatial search techniques in terms of query speed vs. accuracy are evaluated and presented in this paper.