As we’ve just released the first version of our SDK, I would like to give some more details on how it has been packaged and how easy it is to plug it into your very next iPhone application.

We thought that integrating image recognition into your app should be as simple as pluging in any other view controller. In other words you shouldn’t have to care for messy details such as HTTP handlers, activity indicator, portrait vs landscape image orientation, response parsing, etc. That’s why one of our primary design decisions was to package all this stuff into an easy-to-use high-level UI component. If you know how to use the iOS UIImagePickerController then you’re ready-to-go!

So the SDK works as follow:

• let’s use a camera button or any other UI element of your choice
• present the MImagePickerController as soon as this action is being fired
• snap a picture and let’s the magic happens!

If matches are found the SDK returns the list of item IDs corresponding to the data you’ve indexed with Moodstocks API. It’s that easy! Note that at this step it’s up to you to fetch the metadata that corresponds to the recognized items, and display it the way you want.

You’ll find an example application into the SDK (aka DemoApp). This demo app recaps how to configure the SDK, plug the picker and get the results.

This basic app acts as a viewer over the recognition results returned by the SDK: the status, message and list of recognized item IDs are directly printed out on-screen. With Moodstocks API you were able to use your favorite browser as a client to perform some recognition tests via the convenient image_url parameter. Now, with both the SDK and the DemoApp, you’re free to perform your tests directly on your iPhone device by taking the query pictures of your choice!