An Application Platform for Wearable Cognitive Assistance

Wearable cognitive assistance applications can provide guidance for many facets of a user’s daily life. This thesis targets the enabling of a new genre of such applications that require both heavy computation and very low response time on inputs from mobile devices.  The core contribution of this thesis is

15-821 Fall 2017

6 course projectsThe Fall 2017 offering of 15-821/18-843 "Mobile and Pervasive Computing" course included several student projects based on cloudlets and wearable cognitive assistance. This is a YouTube playlist with videos of the student projects captured on the final day of class.

Live Synthesis of Vehicle-Sourced Data Over 4G LTE

Accurate, up-to-date maps of transient traffic and hazards are  invaluable to drivers, city managers, and the emerging class of  self-driving vehicles. We present LiveMap, a scalable, automated system  for acquiring, curating, and disseminating detailed, continually-updated  road conditions in a region. LiveMap leverages in-vehicle cameras,  sensors, and processors to crowd-source hazard

An Empirical Study of Latency in an Emerging Class of Edge Computing Applications for Wearable Cognitive Assistance

An emerging class of interactive wearable cognitive assistance  applications is poised to become one of the key demonstrators of edge  computing infrastructure. In this paper, we design seven such  applications and evaluate their performance in terms of latency across a  range of edge computing configurations, mobile hardware, and wireless  networks,

IKEA Stool Assembly: Wearable Cognitive Assistant

IKEA Stool Assembly: Wearable Cognitive AssistantThis Gabriel application was created by Mihir Bala, a talented freshman CS student from the University of Michigan, as an NSF Research Experience for Undergradautes project under the mentorship of Zhuo Chen.   In addition to being another example of a Gabriel application, it offers the

Assisting Users in a World Full of Cameras: A Privacy-aware Infrastructure for Computer Vision Applications

Computer vision based technologies have seen widespread adoption over  the recent years. This use is not limited to the rapid adoption of  facial recognition technology but extends to facial expression  recognition, scene recognition and more. These developments raise  privacy concerns and call for novel solutions to ensure adequate user  awareness,

Edge Computing for Situational Awareness

Situational awareness involves the timely acquisition of knowledge about  real-world events, distillation of those events into higher-level  conceptual constructs, and their synthesis into a coherent  context-sensitive view. We explore how convergent trends in video  sensing, crowd sourcing and edge computing can be harnessed to create a  shared real-time information system

A Scalable and Privacy-Aware IoT Service for Live Video Analytics

We present OpenFace, our new open-source face recognition system that  approaches state-of-the-art accuracy. Integrating OpenFace with  inter-frame tracking, we build RTFace, a mechanism for denaturing video  streams that selectively blurs faces according to specified policies at  full frame rates. This enables privacy management for live video  analytics while providing a

Cloudlet-based Just-in-Time Indexing of IoT Video

As video cameras proliferate, the ability to scalably capture and search their data becomes important. Scalability is improved by performing video analytics on cloudlets at the edge of the Internet, and only shipping extracted index information and meta-data to the cloud. In this setting, we describe interactive data exploration (IDE)