Browse Prior Art Database

System and method for selective backup of media Disclosure Number: IPCOM000234837D
Publication Date: 2014-Feb-10
Document File: 1 page(s) / 35K

Publishing Venue

The Prior Art Database


More and more people are taking photos and videos on their phone or other portable device, but often this media isn't being backed up. Large corporations (e.g., Apple, Samsung) have the ability to backup everyone's media to a central server, but this isn't feasible for all media, for all users, given limited storage capacities (especially for video), and more importantly, limited local upload bandwidth. This publication describes a solution to this problem, by ranking the media by importance using several techniques, and selectively backing up the most important media first.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 66% of the total text.

Page 01 of 1

System and method for selective backup of media

The problem of someone losing their phone, and losing all their pictures, still exists. It will only be exacerbated by users taking more and more photos and videos. While users may not care about losing a few photos now, likely in 10, 15, or 20 years, they'll want to look back at certain photos, that are more important to them than others.

    This idea uses a statistical ranking method, inferring the importance of certain photos over others, using several techniques such as:

Timing how long media is shown on screen

Whether media contains faces, both familiar and not
Whether media has been shared with others via social networks.

Other inputs to the ranking system are possible. Each photo or video has

"importance" as an associated piece of metadata. Media with the highest importance gets backed up first. The importance could be adjusted using Bayesian statistics (as an example).

Initially, media has a ranking of zero. In one embodiment, when a photo or

video is on screen, the time elapsed is recorded, and this is used to raise the ranking. (A photo shown on screen or longer, or even several different times, is likely to be important to the user.) Further, noting the subjects of the media is a solved problem (either by face recognition, or simply recording the names in metadata by a more manual process). Additionally, if the media is shared, for example by sending it to a social network, its importance ranking will also b...