Ever walked past that airport billboard where you could swear that the model in the picture is looking right at you? Really looking right at you?

Look closer.

That billboard might just be looking at you looking at it!

And even assessing HOW you’re looking at it!

And no, we haven’t been smoking the drapes in the advertising creative department, but advertisers are getting very creative about measuring how we respond to billboards.

You may never look at a billboard the same way again…

Now those cunning advertisers have teamed up with NICTA (formerly National ICT Australia) to get a more accurate measure of exactly how we respond to their advertisements.

Accurately and meaningfully measuring advertising responses has ever been notoriously difficult. SMS messages, 1800 free call numbers, coupons, phone interviews etc can all provide some information, but there has always still been a long way to go.

So How Does This Work?

First of all, a NICTA research team worked with several advertisers to gain a better understanding of exactly what type of information advertisers really needed to better assess the effectiveness of their campaigns.

The NICTA researchers then related this understanding to an existing NICTA project focused on the development of video analysis technology. This software-based technology can scan video footage and extract information based on preset parameters. Specifically the software is able to detect human faces and determine their orientation. By using sophisticated NICTA-developed algorithms, the software can make real-time scans of live video footage and determine and provide accurate data on people’s behaviour and movements.

Results

NICTA embed a small video camera into an advertising billboard. The camera monitors people who come within a few metres of the billboard and the software analyses their responses. Called TABANAR (Targeted Advertising Based on Audience Natural Response), the system can determine whether or not they look at the advertisement and the length of time they spend doing so. Under certain conditions, the software can even determine which areas of an advertisement have succeeded in attracting attention, providing valuable feedback to advertisers on what types of messages and visual elements work.

Through the aggregation of a large number of responses, advertisers can build an accurate picture of how effective their display campaigns have been. If a billboard is able to scroll through multiple advertisements, direct comparisons can even be made of the relative effectiveness of each.

The NICTA research team is now focused on enhancing the video analysis software to allow it to gather even more information. Future versions will be able to recognise whether people looking at the advertisements are male or female, their approximate age and other demographic information.

However there are some limitations. The system is better suited to internal applications because of lighting and weather considerations, and also to situations where the audience is closer (e.g. one or two metres) rather than further away, so that a one-on-one situation can be established, rather than the software having to pick faces out of a moving crowd of people.

This means that the technology is particularly suited to in-store and other internal applications, where particular viewer responses could for example, trigger other messages or offers appropriate to the analysed responses of the viewer.

Commercialisation Opportunities

Early feedback from advertisers has been particularly strong. NICTA is working with a number of parties to implement a small-scale trial of the technology to assess its effectiveness in real-world environments.

The development presents several potential business models for the future. The video analysis software could be packaged with cameras and sold directly to billboard or advertising screen owners. The information gathered could then be provided to advertisers as a value-added service. Alternatively, a company could be established that provided the system as a managed service, delivering information to interested advertisers. The service could be extended to manage sequences of immediate, on location responses.

Future versions of the system could also be used to analyse television-based advertisements, opening up another large market for the technology. Potential also exists within the computer gaming industry, where the technology could be used to improve the interactivity of video games.

For further information:
http://www.nicta.com.au

Dr Chris Nicol presents the 2007 Innovation Lecture in Melbourne, Brisbane and Sydney,
5–7 June.  Click here for information and here to register.

Disclaimer: The Warren Centre publishes articles relating to new technology and innovation that are often based on information supplied by third parties. While an editorial process is applied, we make no exhaustive investigation into the accuracy of the information, thus no liability will be accepted for its accuracy. Please note that in providing this information, The Warren Centre is not supporting or promoting any technology or company, merely seeking to inform. Interested readers should take their own steps to verify the information prior to relying on it in any way.

Innovative Australian Engineering Magicians to Take on the World

Energy Efficient Australian CV Joint Eyes Huge International Markets
Here's Looking at You
"Born Globals" – Business Going Global from Day 1
A "Year of Ideas" for Richer Engineering Returns

CSIRO Robotic Submarine "Starbug" set to transform environmental monitoring
The Warren Centre thanks all its volunteers

Click here to go back to The Warren Centre Bulletin Page.

(c) The Warren Centre for Advanced Engineering, 2007