This API has limited language support. Please do include more languages support as well. Or please share the idea of creating our own custom language train models and use them as our source language.

Can handwriting(Japanese) be supported?

It's not so hard to add Bulgarian language once you have Russian (cyrillic).

Artistic fonts like the ones found in comics are really hard to recognize, especially for North East Asian languages.

Fix that, Google equivalent ocr option handles it a ton better.

Fractions are not being recognized..
many different combinations and characters attempted. Probably similar to symbols issue. Alternative-- How can I train a model to deal with this? I am previewing Form Recognizer but suspect similar results.

Currently vision API is not able to handle ₹ (INR) symbol.
It converts it to 3. So if you want to detect amount ₹ 49, you will get 349 as output.

Hello,

i'm using the handwriting recognition of the Computer Vision API to describe certain german documents. therefore i'm in need of the umlaute aswell as the ß sign. are there any plans on supporting this? sadly the localizable ocr results are not accurate enough.

Regards

When i try to detect license number from image of pattern 12/234232/3454, the slash symbol is recognized as 1.

People need to know what you are doing with their images during and after processing. Please have a clearly stated policy on this.

i want to filter the search about a object or something in the image

Photo of landmark in Vancouver, Canada is incorrectly identified as "a plane parked on the side of the road" - it should at least be identified as a monument or sculpture.

Please provide the taxonomy for tagging and taxonomy used for object detection in the documentation.

It would fine if it is not an exhaustive list. This taxonomy would be more helpful to users to set predefined conditions for objects or tags.

Almost all tools that can generate API Client out of Swagger API defenition, will use OperationId to generate method name in a client. In your swagger API, OperationIds looks like "56f91f2e778daf14a499e1fa", and in C# it will produce method name public System.Threading.Tasks.Task 56f91f2e778daf14a499e1faAsync();
It's extremely hard to recognize what method doing, based on this operationID

Please, modify operationIds in a swagger documentation https://westus.dev.cognitive.microsoft.com/docs/services/5adf991815e1060e6355ad44/export?DocumentFormat=Swagger&ApiName=Computer%20Vision%20API%20-%20v2.0 to make it human and code generator tools friendly.

P.S. I got this swagger file from https://westus.dev.cognitive.microsoft.com/docs/services/5adf991815e1060e6355ad44/operations/587f2c6a154055056008f200

When analyzing regular pics vs. Pics with Halograms embedded inside

Function to search in whole Blob with Images to recognize duplicated images.

Can the API detect humans even from a back view. It might not be very accurate just like humans but after you have lived with someone else for a long time there is a high chance that you can guess who they are even from a back view. Therefore was wondering if the computer detected multiple images of people even from back view whether it would be able to guess.

The guide for recognizeText is not available for php. So I would like to add one. How should I add?