One will come across various types of search engines which specialize in a certain specific area. While some of these are useful, others do not perform up to the mark. However, TinEye certainly does not fall in the latter category. TinEye is an image search engine which allows the user to provide an image URL or upload an image in order to locate where that particular picture lies on the internet. The concept is marvelous, and TinEye performs a great job even if it still lies in a private beta.

On the main site of TinEye, you can go through the video demonstration of how an image is to be uploaded and then searched for to know its different locations on the web. It uses pattern recognition algorithms to find such locations of the images. The result displays a list of places where the picture appeared or was used.

The interesting thing to note is that TinEye does not make use of watermarks or metadata to locate the image places. Rather, it instantly analyzes the user’s query image to create a compact digital signature or ‘fingerprint’ for it. Thereafter, it searches for the image on the web by comparing its fingerprint with that of every other single image in the TinEye search index.

In order to make the service available without causing trouble to the user, TinEye has come up with a Firefox Extension, which requires the user to right-click on the image anywhere on the web by using Firefox to perform the search. Another surprising thing is that the image search engine does not only display the exact replica of the image, but as well the edits and variations of the same picture. It has an index of about 487 million images and further plans to add billions more in the near future.

This service allows users, including graphic artists to find where their work has been used without their permission. It can also help companies who deal in original pictures as part of its business. Not surprisingly, Digg is one of the companies that make use of TinEye’s services to identify websites that have duplicate story submissions with an image.

The only drawback with TinEye is that you need an invite so as to start using its services. The search engine has already reached the maximum capacity for the time being but sill accepts requests for an invite.

Now, it will be interesting to watch what Google does with this remarkable image search engine.