Azure Computer Vision - Legacy OCR and Read (OCR) APIs

Ashkan Izadpanah 21 Reputation points
2022-11-14T19:54:18.143+00:00

We are currently using Computer vision version 2.0 with the following 2 endpoints in our enterprise application,
<endpoint url>/vision/v2.0/analyze
<endpoint url>/vision/v2.0/ocr --> (Legacy OCR API)

Since Legacy OCR API is not going to be supported anymore, we are planning to upgrade to either version 3.2 or version 4 (once it becomes available).

In version 3.x, Async Read API supports both Images and Document (text-heavy) OCR.
As announced by Azure, in v4.0, there will be 2 APIs:
ImageAnalytics v4.0 for Image OCR (general and in-wild images) - sync
FormRecognizer for document OCR (Digital and scanned documents) - async
https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-ocr

Note: we are not currently using FormRecognizer service in our application, and we prefer not to use it (if possible).

Here are my questions:
1- Legacy OCR API is still active (v2.x and v3.x endpoints are still functioning), but Azure is mentioning that this API is no longer supported. When will this legacy API be retiring?

2- For ImageAnalytics v4.0,
a) When will it be available (now in preview)?
b) Will legacy OCR API be available till then?

3- For version 4.0,
a) Other than FormRecognizer service, is there another option for document OCR?
d) Will Async Read API be available or be discontinued?
in other words, will the following Read API endpoints be active?
<endpoint url>/vision/v4.0/read/analyze​
<endpoint url>/vision/v4.0/read/analyzeResults/<operation#>

I appreciate your feedback on this matter since our decision plan is highly dependent on Azure team's feedback.

Azure Computer Vision
Azure Computer Vision
An Azure artificial intelligence service that analyzes content in images and video.
347 questions
{count} votes