Salesforce is ready to join the computer vision business. The company has a new custom image recognition tool.
According to Computerworld, Salesforce announced on Tuesday, Mar. 7, that features of its Einstein system based on machine learning are now live. However, the Vision module will be only made generally available in a couple of weeks.
Einstein Vision allows users to upload various sets of images that could be classified in a series of categories. Based on machine-learning technology, the system will create a recognizer able to identify future images fed into it.
Salesforce first announced Einstein last year. The system's capabilities have been demonstrated at company's Dreamforce conference. Einstein might give Salesforce an edge against the heavy competition against the likes of Oracle and Microsoft.
Customers will be able to use the Einstein Vision APIs in order to create applications that know how to spot pitched roofs versus flat ones or recognize different brands in photographs. The custom recognizer could be integrated by developers into other applications.
According to Salesforce's blog, Einstein Vision's powerful new APIs allow Heroku and force.com developers of all skill levels to build AI-powered apps fast and bring image recognition to CRM. The pre-trained image classifiers and the custom classifiers can be trained by developers to solve a vast array of image-recognition use cases, empowering end users across marketing, service, and sales to be more predictive.
Salesforce will make Einstein Vision generally available worldwide on Mar. 14. The application is a result of last year's acquisition of MetaMind. Salesforce's vision capabilities are similar to IBM's Watson Visual Recognition Service and other competitor machine-vision APIs. In fact, Salesforce also has a partnership with IBM to connect its CRM with Watson.
It is unlikely that Salesforce will be competing with A.I. services offered by other cloud companies like Microsoft, Google, and IBM. However, the company may have a significant role to play in the A.I. ecosystem.
