News

Thank you to the over 140 undergraduates who applied to ODCSSS 2009 from around the world.

Project 1506-dcu: sensecam image gathering

Overview

The SenseCam is a wearable camera which takes up to 2,500 pictures per day and is produced in limited quantities as a demonstrator kit for donation only, by Microsoft Research. It has been developed as part of the "MyLifeBits" project, and is one of the devices used to record one's own personal digital memory or archive. It is designed to record a complete day's activities, visually, by generating masses of photos, storing them as part of a personal digital memory and then using them for some as yet unspecified application. It is one of the family of devices which continuously monitors the wearer's life records the life as photos rather than video, or audio. The SenseCam has a built-in accelerometer to help guide at which precise moment pictures should be taken, i.e. don't take second or subsequent photos if the wearer is not moving. The next generation SenseCams will have support for audio recording, and for GPS tagging of where photos were taken.

As part of a project part-funded by Microsoft Research in Redmond, US, Microsoft Research are donating two of these SenseCams to our lab in Spring 2006. This purpose of this internship is to assist in the project we have, which is to generate landmark images from large collections of SenseCam images. Specifically the intern will develop a software system to manage large numbers of gathered sensecam images, as several hundreds of thousands of images will be collected by two SenseCams in a single year.

Our research lab has extensive experience in managing very large video archives of several hundreds of hours of video content, and has experience of managing photo archives of tens of thousand of images, and web collections of dozens of millions of web pages, but SenseCam will potentially generate hundred of thousands of images per year. While we have ways of analysing and indexing these we do not have a software infrastructure in place to perform the database management aspects of this and this is what is proposed here.

Relevance of Project to the Host Laboratories:

The SenseCam project is one of several projects within the SFI-funded Adaptive Information Cluster which aim to capture a person's life experiences. Others include using a range of biometric signal monitoring devices as well as several "media" projects to record all the TV, movies, etc. one listens to or views during the day. The project proposed here is important because it will enable us to manage very large archives of images taken using SenseCam.

Supervisors:

Prof. Alan Smeaton (AIC, DCU)

Keywords:

Image Processing, Digital Cameras, Image Database

Links:

 
 

http://research.microsoft.com/sendev/project_sensecam.aspx