Google has launched the Cloud Vision API, a new tool that helps developers classify images based on their content. Using machine learning and a REST API, Cloud Vision will separate pictures into a wide variety of categories, metadata can be added, or offensive content moderated out.
The feature packed API also includes:
- Language and character recognition
- Google SafeSearch integration
- Face detection
- Logo detection
- Landmark detection
- Main content recognition for fast and accurate categorization
To show how Cloud Vision API will work, Google produced a short video demonstrating it inside a simple robot, built with the Raspberry Pi platform.
Google demonstrates how the Cloud Vision API can be used
If you’re keen to experiment with the new Cloud Vision API, it’s available as a limited beta release through Google right now, and you can sign up for access here.