Isn't it strange that, although the web is such a visual environment, we can still only search for things with text? Well, not any more.

By applying technical breakthroughs in image search and retrieval, PIXSTA allows web users to browse and look for images using other images - not words – as their search term. Images are retrieved based on their visual similarity: by colour, texture and shape.

The first commercial application of the new technology is the PIXSTA AdImages Network, an online advertising model which allows any fashion retailer to feature products on the shopping sites of PIXSTA’s media partners. Shoppers can browse thousands of products visually, delivering a more intuitive and satisfying consumer experience.

‘A picture is worth a thousand words’? You bet. Well, maybe not that much, but to advertisers the value of a PIXSTA AdImage click-through is significantly higher than any adword – with conversion rates to prove it. As a result, the PIXSTA AdImages network already has pilot customers across Europe.

To media owners, PIXSTA AdImages provides relevant advertising that looks and behaves like content, attracting site visitors. And, for those visitors, there's no more clicking through on adwords to distracting, disappointing or irrelevant products, because they've already seen where they're going.

About the Speaker

Daniel Heesch obtained his PhD in Computer Science from Imperial College London after gaining first degrees in Biology (Oxford, St John's College) and in Mathematics (Open University). He spent two years as a Post-Doc in computer vision with Maria Petrou in the Communications and Signal Processing group at Imperial College London before setting up Pixsta, a visual search start-up based in London."