Akiwi is a new website designed to help you easily find pictures associated with words-wrenches. It is based on a semi-automatic image tagging system that is easier than manual tagging and more accurate than automatic image recognition.
The site was created as a student project at the university HTW Berlin. It uses a collection of 22 million photos and can even suggest keywords on unknown photos uploaded by them users.
Additionally, the site recognizes the basic colors of your photo, which they consider to be tagged, and suggests similar photography in the same basic colors. This is a good search if you want to find photos, say about a website you make, that suits yours.
On its home page site Akiwi Asks you to upload a photo. Then the website will analyze your photo and put keyword on it. At the same time, 10 will suggest other photos that have corresponding keywords. You have the right to set your own additional keywords. and from there you can browse groups of photos until you find the one you are interested in.
In our test the Akiwi website gave us the following results:
Κάτω από τις αρχικά προτεινόμενες από το site φωτογραφίες μπορείτε να δείτε τις λέξεις - κλειδιά που αναγνώρισε η ιστοσελίδα. Μπορείτε να μαρκάρετε μία ή περισσότερες λέξεις ώστε να περιορίσετε τη αναζήτηση. Κάτω δεξιά βλέπετε και το πεδίο όπου μπορείτε να ορίσετε την δική σας λέξη κλειδί για την φωτογραφία σας.
Notice that the site has suggested photos with the same basic colors as ours.
Pressing it key Finalize it is declared that you have found what you wanted and thus the website refers you to the possibility to copy the selected photo(s) to the Clipboard.
The Akiwi website according to its makers is still in progress, and has just been updated in the 1.1 version.