Microsoft

Computer Vision API

Welcome to the Computer Vision API Forum

Categories

API – Any ideas or feedback pertaining to features or enhancements to Computer Vision API.

Documentation – Any ideas or suggestions for the API Reference or Documentation.

Language Support – Submit a request to have a particular language supported.

Samples & SDK Request – Let us know if you would like to see a Code sample or SDK provided.

Custom/Sample Images – Have an image you’ve tested and not getting the results you are seeking? Upload the image and describe the information or tags you would like to be included.

How can we improve Computer Vision API?

(thinking…)

Enter your idea and we'll search to see if someone has already suggested it.

If a similar idea already exists, you can support and comment on it.

If it doesn't exist, you can post your idea so others can support it.

Enter your idea and we'll search to see if someone has already suggested it.

  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. Single Character recognition

    All I'm trying to recognize is single letters. Such as S, M, L, XL and so forth. Why is Microsoft's OCR unable to recognize these characters?

    8 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Text Recognition  ·  Flag idea as inappropriate…  ·  Admin →
  2. Set image compression as a parameter for "Get thumbnail"

    The thumbnail API is excellent, but ships back a highly compressed image.

    It would be great if compression was a variable I could set. As it stands, it compresses the images too much to be usable for what I'm doing with it.

    Imagine you have a 1024x1024 image and you want a 1024x400 crop.

    Imagine the image is mostly blue sky with a plane flying somewhere in the image.

    The API is pretty good at cropping out all the blue sky and finding the plane. So despite the primary target being thumbnails, acting as an intelligent crop tool works pretty…

    10 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  2 comments  ·  Thumbnail Generation  ·  Flag idea as inappropriate…  ·  Admin →
  3. Logo detection

    Is there something on the roadmap for the vision API to include Logo Detection, similar to what Google is offering? https://cloud.google.com/vision/

    8 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Object Detection  ·  Flag idea as inappropriate…  ·  Admin →
  4. Handwriting recognition through vision api

    The ability to recognize the handwritings is needed. For example, the user will write something in their Tablets and my app will convert those into characters automatically while writing.
    It will be helpful for old fashioned guys who don't wan to change their habit but want to be more productive.

    9 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Text Recognition  ·  Flag idea as inappropriate…  ·  Admin →
    Under Review  ·  Raymond responded

    Thank you for your feedback. How do you feel about including cursive handwriting?

  5. Returns location of objects on the image (people, tree, car's, etc.)

    Computer Vision API returns recognized objects locations on the image, what is the pixel location/rectangle.
    Would be of huge use!

    22 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  5 comments  ·  Object Detection  ·  Flag idea as inappropriate…  ·  Admin →
  6. Support data URIs in image APIs (vision/emotion/face)

    For JavaScript clients, there are many circumstances when you want the image as represented in a Canvas to be uploaded to one of the MCS APIs that accept images. The problem right now is that Canvas returns its content in the form of a data URI (RFC 2397), and callers need to convert this to a blob before uploading. It would be convenient if the APIs simply accepted the data URI directly, although it would mean the network payload is larger than need be (because of the base64 encoding.)

    8 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  1 comment  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  7. Vision API: Provide an SDK that does image binarization step at the client

    For most computer vision applications, the first step is to remove color information and binarize the input image. We are proposing, to have Microsoft release an SDK (python or javascript or iOS or Android or others) that performs the binarization part at the client. This significantly saves the amount of data transferred to the server. It can reduce the amount of data pushed to server as much as 12x and sill have no impact on recognition quality. In addition, its savings for microsoft in incoming bandwidth and CPU time spent. Its also a win for the clients where they have…

    5 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Samples & SDK Request  ·  Flag idea as inappropriate…  ·  Admin →
  8. Vision or sound api's for monitoring and publishing ocean wave engeries and spectrums at popular city beaches or world pro surfing venues.

    Vision and/or sound api's for monitoring and publishing ocean wave energies and spectrums at popular city beaches or world pro surfing venues. Could be used in conjunction with NOAA oceanographic services and raspery pi weather stations to provide realtime 'current' conditions at major city beaches or ocean facing ports. Vision api's might be able to provide local government services with usage statistics of ocean resources such as the number of people entering or leaving the water, the number of people on a beach or attending a surf carnival.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Image Analysis  ·  Flag idea as inappropriate…  ·  Admin →
  9. Add the ability to get localized language support for image description with Vision API

    In order to allow a broader use of the service, it would be great to have the ability to specify the language in which we want to get the image description via Vision API.

    49 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    4 comments  ·  Language Support  ·  Flag idea as inappropriate…  ·  Admin →
    Under Review  ·  Raymond responded

    We are researching options to bring this capability to Computer Vision.

  10. Allow control over minimal spacing between words in recognized text

    Sometimes it now recognizes spaces that arren't there, like "Company" is recongnized as "Co" "pa" "ny". It would be nice to indicate the minimal space between word regions, it the regions are very close together it can the merge then and return a single word.

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Text Recognition  ·  Flag idea as inappropriate…  ·  Admin →
  11. OCR returns Chinese Traditional instead of Simplified

    I tested "Read text in images", or named OCR, and found this.
    I selected Chinese Simplified, but I got Chinese Traditional character.
    Please see the fourth char.

    6 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Language Support  ·  Flag idea as inappropriate…  ·  Admin →
    Requesting Examples  ·  Raymond responded

    Thank you for the feedback. The 21 languages supported by OCR are Chinese Simplified, Chinese Traditional, Czech, Danish, Dutch, English, Finnish, French, German, Greek, Hungarian, Italian, Japanese, Korean, Norwegian, Polish, Portuguese, Russian, Spanish, Swedish, and Turkish.

    Have you tested other images using OCR that are returning Chinese Traditional? Please provide other examples where this occurs. These can be added here:

    https://cognitive.uservoice.com/forums/430309-computer-vision/filters/new
    —-Please select “Custom/Sample image”

  12. Product Recognition with Vision API

    The ability to recognize products could enable the following Retail Scenarios:
    1. Build Shopping Lists by taking pictures of products
    2. Detect On Shelf Availability
    3. Detect Misplaced Products on Shelves
    4. Comparison Shopping
    5. Contextual Shopping (Shazaam for shopping)

    8 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Object Detection  ·  Flag idea as inappropriate…  ·  Admin →
  13. Generate proportional thumbnails

    It may be nice to have a feature whereby the generated thumbnail retained the proportionality of the original. In this mode the width and height parameters would be interpreted as not-to-exceed values.

    6 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  2 comments  ·  Thumbnail Generation  ·  Flag idea as inappropriate…  ·  Admin →
  14. Provide a way to run the computer vision api on our own server

    Looking for a way to run an instance of the computer vision API on our own servers

    17 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  On-Premises Solution  ·  Flag idea as inappropriate…  ·  Admin →
  15. Train the vision api for Automatic Number License Plate Plate Recognition (ANPR)

    Current API Vision don't do well with license plantes. This capability will enable lots of scenarios, for example: Customer segmentation in highways, customized parking Customer Experience, etc

    113 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    19 comments  ·  Text Recognition  ·  Flag idea as inappropriate…  ·  Admin →

    The Vision API offers good text-detection with OCR, but it is not currently optimized for license plates.
    We are constantly trying to improve our services and have added OCR for auto license plate recognition to our list of feature requests.

    Thanks

  16. Quality setting on smart thumbnail generation

    I would like to be able to control the quality of jpg output from the "generate a thumbnail" API

    I want to use the generated thumbnails on mobile with the smart crop option as it prevents crops which just centre a image and often chop the subjects in the photos head off but the returned quality of the jpg is by far much lower than would I would like to use within a app (which has a high DPI)

    I guess a 0-100 quality setting which jpgs often provide would be great

    14 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  11 comments  ·  Thumbnail Generation  ·  Flag idea as inappropriate…  ·  Admin →
  17. Fraud detection: finding out if the photo has been photoshopped

    I'm working on a case to detect if a picture has been tampered with, mainly in the field of car Insurance, but could be applied to different fields. Are there any pointers on this?

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Image Analysis  ·  Flag idea as inappropriate…  ·  Admin →
  18. Photo geotagging

    Wouldn't be cool to have an API Function to recognize a building and returning its location ?

    7 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    Under Review  ·  0 comments  ·  Object Detection  ·  Flag idea as inappropriate…  ·  Admin →
1 2 3 4 6 Next →
  • Don't see your idea?

Feedback and Knowledge Base