AIMS: An Immersidata Management System

We introduce a system to address the challenges involved in managing the multidimensional sensor data streams generated within immersive environments. We call this data type, immersidata, which is de ned as the data acquired from a user's interactions with an immersive environment. Management of immersidata is challenging because they are: 1) multidimensional, 2) spatio-temporal, 3) continuous data streams (CDS), 4) large in size and bandwidth requirements, and 5) noisy. By focusing on two speci c applications, Attention De cit Hyperactivity Disorder (ADHD) diagnosis and American Sign Language (ASL) recognition, we propose to study the challenges of two main modes of operations on immersidata: o -line and online query and analysis. In addition, we propose complementary approaches for e cient acquisition and storage of immersidata. The core promising idea behind our proposed approaches is a `database friendly' utilization of linear algebraic transformations on both data sets ...

• Type: Conference
• Year: 2003
• Where: CIDR
• Authors: Cyrus Shahabi
• Keywords: CIDR 2003, CIDR 2007, Data Streams, Immersive Environment, Multidimensional Sensor Data