July 5, 2017 | Written by: Kevin Gong
Categorized: Watson | What's New
Share this post:
A core feature of Visual Recognition since launch, the general tagging capability within the API enables users to quickly and accurately understand the contents of images. Since our major update for the general tagging feature in December 2016, both customers and third parties have reported the capability to be a market leader in terms of accuracy. As part of our continued dedication to provide a better image recognition service that identifies more things with greater accuracy, we’re excited to announce additional enhancements for the general tagging feature in Visual Recognition.
We’ve expanded both the breadth and depth of our general tagging vocabulary by nearly doubling the number of training images we use for the feature. As a result, the Watson Visual Recognition service has greatly expanded its ability to accurately recognize the overall ‘scene’ of an image, allowing it to more completely label the who, what, and where of each image, as well as the activities and dominant colors present. This recognition notably includes some new tags and categories that we now support – such as ‘fireworks’ – so don’t hesitate to test out your 4th of July photos in our demo and tooling!
With our latest update, we’re also providing additional language support for our general tagging feature. In addition to the four languages supported today (English, Spanish, Arabic, and Japanese), the feature will now support three new languages: Korean, Italian and German. You can find details on how to specify language output in our API reference.
Need help or have questions?
We’re excited to see what you build with Watson Visual Recognition, and we’re happy to help you along the way. Try the general tagging feature, read the documentation, and share any questions or comments you have on our developerWorks forums.