Multicolour Search Lab

Colour vision from Idée!

Today we launch the Idée Multicolour Search Lab. In this lab you can search Alamy’s 2.8 million images using multicolours, yes all the colours of the rainbow if you want because when we say multicolour we mean exactly that. Select one colour or several to find images that contain those colours in real time. Everything is better in colour.

redredgreensmall.jpg

For the visual and image search fanatics out there, we have taken our algorithms to the next level: our search results don’t require an exact colour match and cope well with images that contain multiple colours going beyond simple colour histogram matching.

By allowing you to select multiple colours we give you the ability to search for images that match a palette of colours. Want an image with lots of green and just some red? No problem.

Want rainbows? No problem.

And of course none of this would be relevant if it was not for performance and speed! Our super fast visual search engine performs a live colour search in this set of 2.8 million images in a fraction of a second.

So go play and ping us with your feedback. Comment below. Happy searching!

Frequently Asked Questions

Q. What is Visual Search?
A. Visual search is a way to search through a collection of images by visual attributes like shape, colour, texture, etc.. Visual search does not rely on the use of keywords, watermarking or any other metadata–it relies strictly on the appearance of the image.

Q. How does it work?
A. Idée’s Visual Search uses sophisticated algorithms to analyze the visual attributes of an image, including: colours, shapes, textures, luminosity, complexity, objects, regions and many other features. These attributes form a compact digital signature that describes the appearance of each image, and these signatures are calculated by and indexed by our software. When performing a visual search, these signatures are quickly compared by our search engine to return visually similar results.

Q. Can I use the same colour more than once?
A. Absolutely. Using the same colour more than once increases its prominence in the results.

Q. Why do you spell color with a u?
A. Jst becase.

Q. How many images are typically returned?
A. Usually the top 500 images (or 10 pages) of the 3,000,000 or so images we’ve indexed.

Q. Who is Alamy?
A. Alamy is a privately-owned stock photography firm based in the U.K. If you wish to acquire any stock photography for your projects, visit Alamy directly.

These icons link to social bookmarking sites where readers can share and discover new web pages.
  • Digg
  • del.icio.us
  • StumbleUpon
  • SphereIt
  • Reddit

BROWSE / IN TIMELINE
« Idée Labs
» Visual Search Contest Winners!

BROWSE / IN Idée Labs Visual Search
« Idée Labs
» Visual Search Contest Winners!

COMMENTS / 10 COMMENTS

jimsotonna…

jimsotonna dropped by…

jimsotonna Mar 03 08 at 2:08 am

Genius!

Ryan Oct 10 07 at 9:44 pm

Wow. That’s all sorts of awesome.

Alice Oct 11 07 at 11:53 am

This is totally great; I’m majorly infatuated.

Amanda Oct 16 07 at 1:13 pm

Well thank you Amamda! We have yet to recover from your visual search submission!

Leila Boujnane Oct 16 07 at 1:30 pm

sennsationell, I am very inspired

Daniel Nov 12 07 at 3:51 pm

Very cool concept and site. Do you plan on making the images available for download in the future?

Tamar Weinberg Nov 21 07 at 2:49 am

Tamar: thanks for your comment. We are planning on directly linking to the Alamy images. The image used in the multicolour lab are courtesy of Alamy.com. Alamy is a stock photography company in the UK. Their images are available for downloading from their website. In the next release user will be able to click on the result images and see them directly on the Alamy website.

Leila Boujnane Nov 25 07 at 8:10 pm

Hey,

Thanks and compliments on your very nice utility. I’m the author of http://www.colr.org. I just added links from colors and schemes on colr.org to the multicolour search page.

If for some reason you’d prefer these links not exist, please drop a line.

Cheers,

- Lloyd

Lloyd Dalton Nov 30 07 at 2:40 am

The above rave reviews are seconded - when can we expect to see this capability ported to the desktop so I can index and search my own collection of images?

Ben Dec 03 07 at 7:52 am

SPEAK / ADD YOUR COMMENT
Comments are moderated.

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>