AI Builder February update

Nicolas Pouhaer, Senior Program Manager Lead, 2020년 2월 22일 토요일

The February update to AI Builder adds even more capabilities for makers, including a simplified Object Detection wizard, additional fields recognized with Business Card Reader, improved AI Builder controls in Power Apps, and new Power Automate templates to instantly use your AI Builder models.

Manually enter objects for Object Detection

You can now enter object names manually to train your object detection models. This mode of entry is shown by default, but you can still use your object names in Common Data Service.

New fields recognized with Business Card Reader

Updated on 02/25/2020: New Business Card Reader fields will be available in April.

New fields are now available in Business Card Reader. You can extract the Post Office Box and the State fields from addresses detected on business cards in Power Apps and Power Automate.

Updates to AI Builder in the Power Apps canvas studio

Updated on 02/25/2020: Some of the Power Apps canvas studio improvements will be available in April.

Styling and control reset

You can now style all AI Builder components in your canvas apps using the color and border pickers, as well as configure the text properties easily in the property panel. The components also implement the Reset action, which allows you to clean up images and output properties on demand.

Interact with detected text in Text Recognizer

The Text Recognizer component in Power Apps now surfaces a new output property called Selected. You can interact with the component by clicking on detected text to select them. We have also renamed the output property containing the list of text to Results. You can also find more information about the text detected in the image.

Confidence scores and bounding boxes in Form Processor

The confidence scores of selected form fields are now available in Form Processor. In the Results output property, each field and table element has a corresponding key, value, confidence score, and bounding box. You can continue to leverage the fields and tables without detailed information by using the Fields and Tables output properties.

Confidence scores and bounding boxes in Object Detector

You can now see detailed information about all objects detected in your images. For each object detected in an image, you can see the object’s label, confidence score, and bounding box information. This detailed information can be found in the Results property. You can also find the count of objects for each label with the GroupedResults property.

Updated on 02/25/2020: The Power Apps canvas studio improvements will be available in April.

Introducing AI Builder flow templates

To make it easier to use your AI models in Power Automate, we have added new AI Builder flow templates that you can start using today!

AI Builder flow templates

Click here to view them all.

Template	Scenario
Create contact in CDS from a business card using AI Builder on button click	Business Card Reader
Update contacts in CDS from OneDrive(Business) records using AI Builder	Business Card Reader
Click a button to read and save information from documents using AI Builder	Form Processing
Click a button to count objects in images using AI Builder	Object Detection
Send a notification with the sentiment of a manager’s email using AI Builder	Sentiment Analysis
Extract text from images using AI Builder Text Recognition on button click	Text Recognition
Make Office 365 email image attachments searchable using AI Builder	Text Recognition

What does this mean?

You can use these templates to create flows using either your own AI model or a prebuilt AI model in just one click. No more reading through lengthy instructions to create your flow step-by-step. Start with a template and customize it to suit your business needs. You can even discover new ways to use AI Builder in Power Automate. For example, updating contacts in Common Data Service with Business Card Reader allows you to use the info in your Dynamics applications and elsewhere in the Power Platform seamlessly.

Related entities and filtering support in Prediction

Prediction now includes support for related entities and filtering. See this blog post for more information.