Keywords Becoming Obsolete?
Recently there’s been a resurgence in the level of interest in workflow, owing in large part, I’m sure, to the release of Adobe Lightroom and Apple Aperture. One of the topics that frequently comes up when I’m talking to photographers about workflow is keywording. Generally they want to know whether they should bother spending time adding keywords to their images. Very often I’ve suggested it probably isn’t worth the time. My perspective on this is that if you organize your images based on the “photo shoot” or trip they were taken on, you’ll likely be able to remember how to find them. Also, I figure you’ll spend far more time actually applying keywords to your images than you’ll ever spend looking for your images (perhaps this isn’t the most sane thought I’ve ever had).
As I’ve spent more time thinking about this, and the challenges of assigning keywords to photographic images, I started pondering the possibility that some day keywords may become obsolete. Maybe we won’t have any good reason to assign keywords to images in the future.
This might sound a little crazy at first, but I think there’s some truth to this. As I continued to think about this, I played devil’s advocate with myself, and tried to think of examples of situations where keywords were necessary. I thought about some of the weird requests I’ve heard from art directors and others looking for images that conveyed something in particular.
“We need something ‘red’”, they might say. Well, that’s easy to analyze automatically. The predominant color in an image isn’t too difficult to calculate.
How about an image of a particular person? Simple. At some point in the future face recognition will get really good, and we’ll be able to point to an image of a person and ask for all images of that person.
Even object recognition will get to the point that we could draw a sketch or provide a sample photo and have our software find matching images for us.
The more I thought about this, the more I concluded that with the right technology it would be possible to completely replace manual keywords with automated keywords and very sophisticated search. To be sure, we’re not there yet. Not even close. But this is an intriguing concept to me. There’s no question in my mind that at some point we won’t need keywords, because technology will provide a better solution.
I guess in the meantime the only question we have to ponder is whether we need to assign keywords to our images in the meantime…
Explore posts in the same categories: Technology, Digital
December 19th, 2007 at 5:00 am
I feel it is worth the effort because I can’t always remember which albatross was flying, since I photographed 2 or 3 different ones that day, or which whale. And the other benefit is that if and when one of my catalogs gets “blown away” I simply re import my image folders and all the key words come back. Also if you work on an image and make another version to put in a different folder, I believe the keywords will go along with the image. So if the folder you put the image is a new folder that goes into a different folder, you can still find the image via the keywords.
December 19th, 2007 at 9:23 am
I’ve had requests for things like “tropical plants” or “flying goose”. It’s going to be a long time before image recognition software will support that level of specificity on a non domain limited set of objects.
Further more, I’m not sure we’ll ever have image recognition software that can tell you if a picture is “melancholy” or is an appropriate picture for metaphors or phrases (picture of businessman standing in line associated with “early bird gets the worm?”)
December 19th, 2007 at 9:29 am
I can think of a few cases of kinds of images where recognitiontechnology probably won’t catch up to keywords.
One is noting species names. I doubt that recognition technology can distinguish between shorebird species, or warbler species, or different skipper (butterfly) species. People have a hard time - I can only imagine the expert system required.
Another is special file attributes. If you keyword a file as “portfolio” you can return a list of those files. You can include this in a file name, but that can complicate backups if file names change.
Another case of special attributes is the file version you prefer to print, if you have multiple similarly named files. You can store files like this in one location - that’s a possibility with portfolio images as well - but you might want more flexibility than that.
And then there’s the question of whether operating system fetch/query technologies will satisfy all the use cases of digitial asset management - I wonder about that.
December 19th, 2007 at 2:43 pm
I tend to think assuming search can’t eventually solve this might not be giving technology enough credit. This will take a very long time, to be sure, but I think it is possible.
Bryn cited the examples of “tropical plants” and “flying goose”, both of which I think, given enough time, could be figured out by image recognition somewhere down the line. And facial recognition of emotions is something that has been done with modest success.
In any event, I think it is a very intriguing topic, and suspect it is something that will get increasing attention as time goes on (lots of time, admittedly).
December 19th, 2007 at 3:29 pm
I agree that technology could definitely eventually distinguish “flying goose” from a non flying goose or a flying crane if somebody set their mind to it (heck they are doing facial recognition already, and we all should know not to underestimate technology), but what I think is the issue, is that technology research is driven by money and who is going to pay to develop software that can distinguish a flying goose from a standing goose?
Casinos have funded a lot of facial recognition work, maybe they will pay for general object recognition. It’s a similar problem to computer assisted diagnosis and target detection, and those have a lot of funding (but not a whole lot results)
All the recognition software will likely be complex and will be run on the photos and tag them accordingly rather than run in real time, so there will still be keywords maybe just auto generated ones.
So I agree with you tim, given enough time and money we won’t need keywords, but one has to ask how long are you going to live?
I think there will be a lot of assistance with keywords.
Automated GPS tagging will help tag with locations (and may be aid object recognition)
Facial recognition will help with who is in the picture (though you will need to train it in the short term — here’s a pic of my dad, find other pics of my dad and tag them)
predominant colors can be done
And image similarity could be done so that it could auto suggest keywords from similar photos in your collection.
Google has an effort to get users to tag photos in their image search, and they are trying to develop algorithms that would apply those tags to other similar images, and thus be searchable, so maybe there is a way of leveraging all the images that are on web pages with descriptive captions, query the db for a similar image (not trying to figure out what it is, just similar) and then apply those “potential keywords” to images so that when you search for images you get exact matches and then maybe matches.
One analog that makes me think this all might be possible is that a lot of people don’t use bookmarks anymore, they just type their query in to google.
Also along this vein, you have people that ask for the image “with the thing running down the middle” which is hard for humans unless they were recently talking about that image. Though john nack of adobe points out a tool that lets you look up chinese characters by drawing them http://blogs.adobe.com/jnack/2007/12/trajan_the_hack.html so maybe you could draw salient portions of the image and find it.
December 24th, 2007 at 8:22 am
Never say never but until that day comes, I will rejoice in the rewards of keywording
January 17th, 2008 at 5:28 pm
Until my workflow is perfected, I NEED keywording. And that may be another week or 2080.