Could we have the API File size limit increased? currently we work with a lot of imagery with high resolution images that average ~8MB.

Can the Size limit be increased

It would be great if we can specify a parameter for OCR to just read printed or handwritten text. In the case of scanned documents it would improve the accuracy and reduce the API's workload.
It would also make it easier to visualize the detection and read/check the accuracy of large batches of tests

I've successfully been able to use vision to extract handwritten text out of fixed forms by knowing the coordinates of each form field. However many of my forms have checkboxes and/or radio buttons that the users will be filling in with pen. It doesn't seem that vision has a way to detect this type of content.

You have two images like tax forms. Compare the two images. Return a confidence they are the same. Return a set of cords for the image where the primary differences occur. PDF as well as jpg and bmp support would be nice.

Firebase started offering offline on-device image recognition.

https://firebase.google.com/products/ml-kit/

Are there any plans to allow use to do something like OCR offline like Google does?

In the OCR API from Computer Vision API - v2.0, for every word I get the text and the bounding rectangle. Would it be posible to also get the font type and optionally font size/other parameters?
Usage scenario: overlay new text over the old one, for example live translate.

My applications allows end-users to upload images that are used on the site. Users do not understand how to upload images with the correct aspect ratio, size, pixel ratio, etc. I need tools to allow me to process these images and correct these issues. I am currently using Cloudinary but would consider this Azure service if you can add these features.

Help me Identify dog or cat by there nose print. it is can be use for other animal not only for Human face or fingerprint.

How to send a local image instead of URL to Microsoft Computer Vision API using JAVA

MS Handwriting recognition is best-in-class in reading poor handwriting, but seems algorithm chooses not to recognize individual symbols or characters by themselves. All text is not written in full words.

The following examples never process the same symbol correctly on successive lines. Symbols, or even "0" or "o" are ignored when left by themselves.

I have trained a project at customvision.ai/projects. How can I migrate / import this to my production Azure subscription?

To disambiguate similar names, include a DBpedia graph id in the response. This will help more with categorizing results and using the data in more meaningful ways.

To avoid homonymy problems, it would be very useful to add a unique identifier to recognized celebrities in the api response

postRequest =(String)("POST ") +"/vision/v1.0/analyze?https://upload.wikimedia.org/wikipedia/commons/1/12/Broadway_and_Times_Square_by_night.jpg HTTP/1.1\n"
+ "visualFeatures: Categories,Description,Color"
+ "language: en\n"
+ "Content-Type: application/json"
+ "Ocp-Apim-Subscription-Key: {mysub key}\n"
+ "Host: westus.api.cognitive.microsoft.com\n"
+ "Content-Length: 152\r\n\r\n";

As shown above, I'll get an error number of 401,which means Access denied due to missing subscription key , how do I organize the content of my post request ?

We have some standard forms that are in handwritten format but the pattern of the text is always the same (I think this can apply to a large number of scenarios). Allowing the handwritten recognizer to be configured with a expected pattern via a simple regex could improve a lot of scenarios.