This paper presents a comprehensive dataset intended to evaluate passive Human Activity Recognition (HAR) and localization techniques with measurements obtained from synchronized Radio-Frequency (RF) devices and vision-based sensors. The dataset consists of RF data including Channel State Information (CSI) extracted from a WiFi Network Interface Card (NIC), Passive WiFi Radar (PWR) built upon a Software Defined Radio (SDR) platform, and Ultra-Wideband (UWB) signals acquired via commercial off-the-shelf hardware. It also consists of vision/Infra-red based data acquired from Kinect sensors. Approximately 8 hours of annotated measurements are provided, which are collected across two rooms from 6 participants performing 6 daily activities. This dataset can be exploited to advance WiFi and vision-based HAR, for example, using pattern recognition, skeletal representation, deep learning algorithms or other novel approaches to accurately recognize human activities. Furthermore, it can potentially be used to passively track a human in an indoor environment. Such datasets are key tools required for the development of new algorithms and methods in the context of smart homes, elderly care, and surveillance applications.

OPERAnet: A Multimodal Activity Recognition Dataset Acquired from Radio Frequency and Vision-based Sensors.

With the development of motion sensing technology, motion sensor based services have been put into a wide range of applications in recent years. Demand of consuming such service on mobile devices has already emerged. However, as most motion sensors are specifically designed for some heavyweight clients such as PCs or game consoles, there are several technical challenges prohibiting motion sensor from being used by lightweight clients such as mobile devices, for example: • There is no direct approach to connect the motion sensor with mobile devices. • Most mobile devices don’t have enough computational power to consume the motion sensor outputs. To address these problems, I have designed and implemented a framework for publishing general motion sensor functionalities as a RESTful web service that is accessible to mobile devices via HTTP connections. In the framework, a pure HTML5 based interface is delivered to the clients to ensure good accessibility, a websocket based data transferring scheme is adopted to guarantee data transferring efficiency, a server side gesture pipeline is proposed to reduce the client side computational burden and a distributed architecture is designed to make the service scalable. Finally, I conducted three experiments to evaluate the framework’s compatibility, scalability and data transferring performance.

Yiding Chai

Papers

A General Framework For Motion Sensor Based Web Services