TinEye is a brand-new image search engine, original and innovative in more than one way and which should interest, even push Google Images and Videos to develop the traditional keyword search for multimedia content…
Leila Boujnane, CEO and French co-founder of the Canadian company Idée Inc., which publishes TinEye, sums up their originality as follows: “To sum up, Idée (TinEye) is to images and videos, what Google is to text (…) you submit an image or video and our engine will find all documents wherever they are, whether they are on the web or in paper documents like newspapers or magazines. »
Offered in a private BETA version and soon in a public version, TinEye notably brings a break in terms of the use of an image search engine, since it is no longer by entering a keyword but by directly submitting an image or its URL, which the engine will search in its database of 500 million images, to return the list of sites where the image in question is found.
TINEYE, HOW DOES IT WORK?
No facial recognition or analysis of the content of an image as specified by the engine “TinEye looks for the specific image you uploaded, not the content of the image. TinEye cannot identify people or objects in an image. “.
The TinEye algorithm is actually based on the analysis of the pixels of an image to identify an image and its characteristics, allowing the engine to find an image even if it has been greatly modified in size or altered in color.
TinEye continually crawls the web for new images and creates a unique digital signature or “fingerprint” for each photo it finds before adding it to its index.
It then only remains for the engine to compare the fingerprint of your image with those of its own database.
WHY USE TINEYE AND REVERSE IMAGE SEARCH?
IMAGE AND COPYRIGHT PROTECTION
With the web, too many users and publishers of websites feel free. Can reverse image search change everything? Content theft is a scourge for content creators today who put in the effort and put significant resources into a content strategy. The protection and management of image rights, which has already interested AFP and Getty Image – customers of the solution, is necessary in order not to see its image and photo content plundered by the competition. But that’s not all since the technology developed by Tineye also allows companies like Digg to be able to quickly identify duplicate photos. The technology could also be of interest to Google News, Wikio, and Flickr…
Tomorrow with the search for videos, which TinEye announces available for the end of the year, the engine could make it possible both to follow the diffusion and the propagation of a viral video, while offering a new solution to monitor and carry out the protection of audiovisual rights on the Internet, and without having to go through cumbersome DRM-type solutions.
Web content publishers will thus see their content “protected” or more easily identifiable on third-party websites. Thus, rich content such as infographics cannot be taken over. Indeed, if the textual content is secured by the “Duplicate content” penalty, this is not the case for the image content. By finding the copiers, you can ask to credit your content and make a backlink on your website: which is positive for your SEO for Franchise! If you have invested in professionally shot images, you secure your media content.
HOW TO DO A REVERSE IMAGE SEARCH WITH TINEYE?
A REVERSE IMAGE SEARCH FROM THE TINEYE SITE
On the TinEye website, you can:
- Upload your image or photo from your computer
- Copy the URL of an image to launch an image search
- Drag & drop an image from your computer
USE A TINEYE EXTENSION
TinEye has a dedicated extension for 3 major web browsers:
- TinEye For Firefox
- TinEye For Google Chrome
- TinEye For Opera
All you have to do is right-click on the image you want to search for matches.
If the base of nearly 500 million images all the same may still seem relatively low, it is already a sufficient base for launching the engine and judging its relevance.
Example with this photo of an amateur photographer friend, of a street in Portugal.
After uploading this photo, TinEye finds it on 5 websites, yet it is a photo released from a single source and little known.
In particular, there is a site which offers language stays (Active Langue’s) and which has taken over, without authorization (it’s not good), this miniature photo to illustrate its products in Portugal.
In general, the tests carried out since the beta version of TinEye are impressive and will continue to gain in relevance as the database of fingerprints of the images will expand, pending the arrival of the TinEye engine. For the video which will end up ensuring him a worldwide success and which already seems to be off to a good start!
To go further and to read in the expansion, the interview with Leila Boujnane , CEO and French co-founder of the Canadian company IDEA Inc. which publishes TinEye and identical image identification solutions or similar.