Augmented Reality is starting to get the hype that social media applications had a few years ago, and widgets a few years before that. I’d say that we’re increasingly being asked about augmented reality, however when you talk about the cost, it is amazing how quickly the interest can die. I’m finding this area fascinating and will be looking more into it (including, one hopes, what the revenue models might be), so here is a little primer on augmented reality (AR) and mobile AR.

Classically, AR happens when the camera on a digital device (attached camera to computer, inbuilt webcam, mobile camera) scans an image which is coded to trigger an overlay of a virtual visual image on top of the original image, thus augmenting it. There are three key elements to Augmented Reality: the image which triggers the augmented reality software (the reality); the software itself which which is able to recognise the image(s) being steamed through the camera lens; and the results of the trigger, which is the augmentation.

OK, the image which makes all this happen is known as the trigger. Triggers tend to come in what are called ‘hard’ forms, such as 2D codes (like QR codes or Spot codes) or in ’soft’ forms which are just general images or pictures. Recognising the image initiates the process and tracking is critical as it allows the image to continue being recognised as to moves around. Hard marks, being explicit, are easier to initiate and track, however any image can be used as a trigger as long as you have some pretty robust image recognition software. Soft marks take longer to set up, to train so that the image can be tracked in different light, slant, colour, proximity etc. Training is necessary in both cases, but it is longer and needs to be much more through where a soft mark and not a hard one is used.

The software which looks for this trigger can be browser based (server side), or client site through something like an Active X control. In either case, you click on a link/application, which grabs contorl of the camera and asked to hold the image up to it. The system looks to recognise and lock in on the image to produce the augmentation on top of this visual which can be seen in the camera feed into the screen.

When software sees a mark it recognises, it plays the video, 3D, animation or whatever which is queued to be triggered by this recognition. The content played can include sound, movement and, in more sophisticated cases, options for interaction (buttons, numbers). In a very neat implementation, US Postal Service provides a 3D image of a postage box, and you can place your goods inside the image to see what size box you need. While a tape measure would do the job, this is much sexier!

I’ll be looking more at a few clever uses of AR and in particular on mobile (here’s one where the mobile becomes a mouse), but a few sites to look at (car related as I was research this topic) are Mini in Germany, which really focussed on the start to finish importance of a strong creative concept, and this neat one where the car ‘explodes’ before you (a definite money shot) for Toyota.

There is a downloadable (Active X) version which ran recently for News Ltd to promote Night at the Museum 2. You can, from this site, download the images, the control and make this work for yourself.

Expect to see a lot more on this, especially on mobile given the perfect location of the camera and the ability to actually locate where the phone is (which, as I’ve mentioned before, really make for a perfect native mobile partnership).