We are using cookies to give you the best experience. You can find out more about which cookies we are using or switch them off in privacy settings.
AcceptPrivacy Settings
N e w ! We are CI Hub reseller. Contact us to see a demo and/or to discuss opportunities.

Computer Vision with Color Factory Enterprise

Improving asset discoverability through automated keywording

FotoWare now employs Azure Cognitive Services to automate keywording of assets so users can discover their content more easily.

Azure Cognitive Services is the collective name for a number of machine learning (aka AI) services, two of which are now made available in Color Factory Enterprise:

1Computer Vision – automatic image recognition of objects (landmarks, buildings etc), animals and famous celebrities

2. Face recognition – a solution that allows you to train your own data model for face recognition and automatically tag subsequently processed images where subjects are recognized. For example, the Face service can be trained to recognize the employees in your company.

It is estimated that as much as 10% of images in a collection become unsearchable because of poor, inconsistent or no keywording, making them impossible to retrieve. Clients are always looking for a way to do more with less. Adding metadata to assets is always the most labor-intensive part of managing a collection of assets, and anything that can be done to improve this process is going to mean saved time, cost and potentially, frustration.

FotoWare’s take on the solution is employing Microsoft Cognitive Services in the Azure network to offer best-in-class automatic image recognition. The engine returns not only keywords to describe objects, but can also produce meaningful image descriptions in several languages. The engine also recognizes famous landmarks and well-known personalities such as actors, politicians and other influential people, which means pictures can be recognized, described with human-readable text and made searchable in the Digital Asset Management system. Image recognition also works well on stock images, so users who rely heavily on purchased stock imagery can benefit greatly from this technology.

Keywords Auto Tagging