TechHui

Hawaiʻi's Technology Community

I just learned about a new image search technology called TinEye and it reminded me of the things I heard about ilovephotos.com so I thought I'd share it here. (*I know very little about both companies so I'm not sure if these two technologies are really comparable)

If you watch the video on the TinEye site, you can see that it is a brand new way to search for images on the net. It doesn't search for images based on tags and file names but rather uses the actual pixels in the image. Pretty cool stuff.

To me it seems like this technology might grow into a great way to find websites that pirate or illegally copy your hard worked designs and illustrations. Oh and also a great way to track down those wild party photos from college that were never meant to be in the general public.

Here is the Japanese article for those interested.


TinEye Website
*They are still in private beta so you have to get the invite to try it out.

Views: 64

Comment

You need to be a member of TechHui to add comments!

Join TechHui

Comment by Daniel Leuck on July 2, 2008 at 6:07pm
Hey Brooke - I really like View Search Hokkaido. It sounds like a fun project!
Comment by Brooke Fujita on June 29, 2008 at 6:53pm
There are government-backed projects in France and Japan pursuing next-generation image searching. Here's the one in Japan: View Search Hokkaido (I developed a web-service driving the back-end).

@Dan
> Given the enormous computing power brought to bear by Google to index HTML,
> I wonder how it would be possible to build a content based image search that covered
> a substantial portion of the web in a cost effective manner.
Good insight, especially the bit about Google.

One of the driving forces behind this View Search Hokkaido and its wider-ranging parent project Information Grand Voyage, is to take on the search giant (singular or plural??) vis a vis image searching. Images can be machine-processed and categorized by n-dimensions (colors, tones, shapes, edges, amount of noise, etc.), and then like images can be grouped by calculating euclidean distance. Very processor intensive, yes, but if these processing tasks can be broken down in just the right way, they can be attacked in parallel by using such libraries as Yahoo!'s Hadoop, or maybe Google's MapReduce. Or you could simply build your own pc cluster (virtual, even) to do the grunt-work.

At least, that's the theory as I grokked it.

My $0.02
Comment by Daniel Leuck on June 29, 2008 at 11:43am
Hey John & Scott,

I got an account on TinEye yesterday and gave it a spin. Per the article Scott referenced, it works well with photos and complex illustrations even when they have minor modifications such as embedded text or certain regions changed. I wonder how well it does with photos that have had subtle filters applied to the entire image.

John: The main challenge we found were (1) resource utilization and (2) flexibility. Not surprisingly these system can consume very serious amounts of computing resources, making the business case for offering such services challenging and often unrealistic.

That makes sense. Given the enormous computing power brought to bear by Google to index HTML, I wonder how it would be possible to build a content based image search that covered a substantial portion of the web in a cost effective manner.
Comment by Scott Murphy on June 29, 2008 at 10:07am
Daniel and John, that is really interesting and thanks for sharing. Maybe the limitations won't allow the technology to be wide spread but I hope to see it grow and be applied to different websites.

I later found this by the same company. I thought it might be a cool idea if someone applied this technology to a dating site where you can narrow users based on similar "looks". It might be applicable for a stock photography site as well.
Comment by John on June 28, 2008 at 11:57am
I did research and testing in this space not too long ago. TinEye is an example of Content Based Image Retrieval and there are a number of companies working on such solutions.

The main challenge we found were (1) resource utilization and (2) flexibility. Not surprisingly these system can consume very serious amounts of computing resources, making the business case for offering such services challenging and often unrealistic.

The flexibility aspect is an issue of how broad a range of image types the system can handle. For instance, Dan mentioned issues with line art. These systems often have issues with different shapes, patterns, color combinations etc. Obviously it depends on how the system is optimized but there are some serious trade-offs in making a general purpose image searching solution.

One UI technique I have seen use is graphic tools that lets the user select a portion of an image for matching. For instance, you might have a picture with a person in yellow raincoat standing next to the golden gate bridge. The UI allows you to mark a rectangle around the person in the raincoat. There are other UI techniques as well but the general point is to allow human intelligence to relive the strain on the underlying matching system.

It's a great concept but will likely work best in narrow, specific applications like product searching such as like.com offers.
Comment by Daniel Leuck on June 28, 2008 at 9:17am
Great find! TinEye looks very interesting. I wonder how they calculate image distances for partial matches and how they index. Akimoto-san's comment that it doesn't work well with logo type line art is interesting. I assume its optimized for photographs and complex illustrations.

Sponsors

web design, web development, localization

© 2019   Created by Daniel Leuck.   Powered by

Badges  |  Report an Issue  |  Terms of Service