I’ve been working on this problem of image retrieval since before digital images became commonly available to the average consumer. Now digital photos are not only available, they’re part of everybody’s lifestyle. And in this time, we’ve not developed any systems that can help organize and retrieve photos. I started doing research in this area in around 1990 and co-founded a company called Virage (that really meant Visual Information Retrieval Age) in 1993.
Â
Today, I have more than 15,000 digital photos (increasing by about 1,000 every other month) on my computer, all in different folders. There’s no system to help me. And I’m getting frustrated with the situation. I need a system that will help me organize my photos. Believe me, I’m not only the person with this situation. This is a serious problem that needs a solution.
Â
I would love to see researchers in the field of multimedia information systems leave their obsession with only using image intensity values and use any and all information that comes from cameras, personal calendars, maps, and any other sources on the Web to help solve the problem.
Today’s search engines—including Google and Yahoo—use only metatdata available on the page on which the image appears. And photo-oriented systems also use very limited, if any, information.
Â
I for one have decided to consider photo retrieval a true multimedia problem, and hence use all the metadata from cameras, calendars, maps, the Web, and also intensity values in images.