Read more: "Lifelogging: This is your life, on the record"
LIFELOGGING is moving beyond just counting how many steps you take. Soon it will encompass almost everything we do, generating vast quantities of data in the process. But how can you organise and browse all the video, audio, image and text data you amass? What we need, says Cathal Gurrin of Dublin City University in Ireland, are ways to automatically annotate files with descriptive, searchable tags and to easily browse them. "With lifelogging, it's like search engines back in 1997," says Gurrin, who has been a lifelogger for seven years. "There's so much that needs to be done."
In Hong Kong, film-maker and artist Alan Kwan has been lifelogging since November 2011 using a video camera fixed to his glasses. He has developed Bad Trip, a 3D virtual world in which people can explore his memories.
"Every night I upload the videos – my virtual memories – to the world in the video game. Different houses store different memories. For instance, the first house is the one that stores my memories with my current girlfriend," says Kwan.
"My secret, intimate or embarrassing memories are stored in flying houses where players cannot gain access – but if they climb hills and get close enough to the flying houses, they can hear whispers of the secret memories."
Making lifelogs game-like may be fun, says Gurrin, but it's a slow way of finding what you want. So he and his colleagues in the Human Media Group at Dublin City University are taking different approaches. He and team member Yang Yang are looking into how logs recorded on smartphones can be browsed and displayed, with more important memories given greater prominence graphically. But automatically deciding which memories are the most important ones is not easy. So Gurrin is working with another colleague, Lijuan Zhou, on a data ranking system called MemoryMesh that uses a technique similar to Google's page rank algorithm to do the job.
"It is a challenge to find out the importance of the data in a big archive," says Gurrin. "A few logs will be unique, novel and interesting and the idea is that the system will rank them and link you to those more easily – they'll just jump out at you."
In the future, Gurrin envisages that wearable headsets incorporating brainwave sensors will do memory ranking. "A spike due to excitement in a certain part of the brain could then be logged alongside the video to infer its importance level," he says.
Fast browsing through memories will be dependent on such annotation, says Gurrin. To do this efficiently, we'll need machine learning algorithms that can automatically recognise and tag all forms of lifelogged data. "Only then will we get the killer applications," Gurrin says.
This article appeared in print under the headline "Revisit your past in a virtual stroll down memory lane"
- New Scientist
- Not just a website!
- Subscribe to New Scientist and get:
- New Scientist magazine delivered every week
- Unlimited online access to articles from over 500 back issues
- Subscribe Now and Save
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.
