An API is a programmatic way of accessing ObjectDescription. This lets you programmatically get a description for an object (such as an image) based on its fingerprint (MD5 checksum). Through the API you can also submit descriptions for images.
Two types of API have been developed. One is based on the HTTP GET and URL query string. The other is based on SOAP.
The APIs are in the final stages of review, and documentation will be published shortly. If you would like to request a specific feature or are interested in early access to the API, please contact us.