Image license retrieval

From Crucible Network Research Projects
Revision as of 13:45, 2 September 2011 by afb21 (talk | contribs) (Created page with "This student design project completed for client John Piper at Kodak European Research explored the use of machine vision technologies to manage image licensing opportuni...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

This student design project completed for client John Piper at Kodak European Research explored the use of machine vision technologies to manage image licensing opportunities within a museum context.

Original design brief: "Smart Poster Picker"

Many people like to buy poster reproductions of famous artworks. This could be done by using a mobile phone to take a picture of the painting they would like to hang on their wall, and sending an MMS message to a poster distribution service. Design a system that retrieves and ranks probable matching candidates from a database of posters, based on a mobile phone image. A simple metric for matching might be the relative averages or variances of hue in each quadrant of the image. Where the photo is not good enough to provide an effective match, users should also be able to navigate through thumbnails of the available posters, arranged according to the match metric for aesthetic graphical navigation.