Cloud-based smart image recognition processing solution
Marketing materials
Visual ID is an SME based in Dublin, providing virtual content production systems for print and digital outputs for the fast-moving grocery market. Their products allow customers to store, collaborate, distribute, produce and personalise their digital marketing materials. These materials can then be deployed to print, social media, digital screens and the web.
Digital asset management
Visual ID’s online digital asset management system stores hundreds of thousands of digital images of products from different brands. The number of images is constantly increasing, as clients upload new images and alterations happen to branding and product specifications, etc. Most images are provided with little or no naming protocols, making them impossible under current norms to search, locate and re-purpose. Manual intervention is required – under the current system – to name and index these images and eliminate duplications and similarities. The costs associated with this manual process are excessive and extremely inefficient.
4,204
active users
55
locations worldwide
120
files created/hour
“The manual processes we used have now been replaced by an automated system which is faster, cheaper and transparently scalable. This system provides the company with a distinct advantage over our competitors in the marketplace. We see this development as a stand-alone commercial opportunity to market it to brand owners and the greater retail market worldwide.”
David McDonnell
Managing Director, Visual ID
The Visual ID & COMMAND partnership
The COMAND Technology Gateway, in cooperation with Visual ID, designed a cloud-based smart image recognition processing solution for Visual ID’s digital asset management system. The solution provides a scalable and efficient system with the ability to use smart machine learning, and identify images by brand and variant by using the newly developed smart image recognition system.
A special learning algorithm has been developed for machine learning of new brands, identification and indexing of near duplicate images, to interpret graphical and normal text and associated with the image and to more readily facilitate service requests to the virtual infrastructure.