Microsoft

Face

Welcome to the Face Forum

Categories

API – Any ideas or feedback pertaining to features or enhancements to Face 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.

                               Attention!





We have moved our Customer Feedback & Ideas for Azure Cognitive Services portal to the Azure Feedback Forum.





Please go to the link below to access our new Feedback and Ideas Page.



  1. Face API - Detection finds nothing BUT 200 Status code returned (204 instead?)

    The Face API sometimes does not detect faces in the source document. It still returns a Status Code of 200.

    The suggestion is that a 204 Status Code, "No Content", be returned instead.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  2. Face API "detection_02" DectectionModel - can produce negative coordinates

    On photos where the face is cut off, the detection_02 model has produced negative coordinates, i.e. "top" and/or "left" rectangle values where the face would be if there were any data there but beyond the actual bounds of the photo.

    If there is no data, the API shouldn't return a rectangle for a non-existent area.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  3. Face API Logic App Connector - Add parameter for DectectionModel

    The addition of this parameter will save time by making both customers and several Microsoft Tech Support personnel aware that there is more than one (1) model.

    Additionally, the LA connector, by default, uses "detection01" whereas the demo page (https://azure.microsoft.com/en-us/services/cognitive-services/face/#demo) uses "detection02".

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  4. exception unhandled

    i copied and pasted the whole code on github then run the program howerver, the program cause an error at DetectFaceExtract function. how I can fix it

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

    We’ll send you updates on this idea

    0 comments  ·  Documentation  ·  Flag idea as inappropriate…  ·  Admin →
  5. Add face landmarks for chin / yaw line.

    Can you please add facial landmarks for the chin of a person. To create biometrical images, the chinline has to be in a specific position within the image.
    See example *9 for usefull landmarks.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  6. What value I can fill on targetFace option?

    Question 1: Maybe 2 sentences are contradictory. Can someone explaine for me?
    "targetFace" rectangle should contain one face. Zero or multiple faces will be regarded as an error.
    If there is more than one face in the image, targetFace is required to specify which face to add. No targetFace means there is only one face detected in the entire image.

    Question 2: What value I can fill on targetFace option?
    targetFace is required to specify which face to add.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  7. How can I find that person and quickly?

    I get a faceId from Face - Detect then I get a personId from Face - Identify. Finaly, I get the name of that person from LargePersonGroup Person - Get. It is a happy case. But in case, I do not know largePersonGroupId and total of my LargepersonGroup and person per LargepersonGroup are maximun. How can I find that person and quickly?

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  8. Face Api Exception

    I am trying to use your sample WPF application. When I choose a local image for detection, I get this annoying exception as below.
    "Object reference not set to an instance of an object."
    I attached the screenshot including stacktrace.
    I suppose there is/are some major issue(s) you need to fix! (DetectAsync(fStream,...)) Thx.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  9. Add migration path for different detection and recognition models

    As it stands currently, it is very cumbersome and onerous to migrate from 1 detection/recognition model to another when we have live customer data already.

    This works easy enough for a new customer as all of their data will be at the detection/recognition model that you specify.

    However, for existing customers, we will have existing facelists and persongroups already with data in it, and short of reprocessing ALL the face data, we don't really have a viable way to migrate to the new models.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  10. face to person api documentation not up to date and is not consistent

    Hello,

    It seems like this documentation(https://docs.microsoft.com/en-us/rest/api/cognitiveservices/face/face/verifyfacetoperson) is not up to date. In the api, it implies that personGroupId or largePersonGroupId is optional, but at least one of them are required. For example,

    When I submit a request without personGroupID or largePersonGroupId, it returns:

    {

    "error": {
    
    "code": "BadArgument",
    "message": "Request body is invalid."
    }

    }
    while verifying FaceToFace(https://docs.microsoft.com/en-us/rest/api/cognitiveservices/face/face/verifyfacetoface)

    you'd only require two face ids and no group ids. I'm not sure where to file this improvement/bug report, so I am posting it here...

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  11. Face Recognition Sample code

    I would like have a sample face recognition code such as that can create the person group and trian the group.
    I always got API error when set up the program by following the tutorial guide

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

    We’ll send you updates on this idea

    1 comment  ·  Samples & SDK requests  ·  Flag idea as inappropriate…  ·  Admin →
  12. Why is there not a chin and/or jawline landmark?

    The whole world is making masks these days. Most facial size difference for masks are measured in S,M,L by length of the face effectively from the bottom of the chin to the nasal root R/L. It is referred to as the Menton-Subnasale Length (lower face length) Reference measurement E in the attached image.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  13. Color schemes

    I find no easy way to identify the skin color of faces. Surprisingly i see attributes for hair color. Any reason why?
    Also, there seems to be no easy way to identify the color of a specific object in object detection too.

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

    We’ll send you updates on this idea

    0 comments  ·  Samples & SDK requests  ·  Flag idea as inappropriate…  ·  Admin →
  14. Need Simple Face Detection / Verification Process

    Existing One:


    • Need to capture image and then send it to web server to make it as URL

    • After that need to get generated URL and send it to FACE DETECT API to generate FACE ID

    • On successful attempt we will get a FACE ID for the uploaded image in response

    • Then need to call FACE VERIFY API with two FACE ID's for verification

    • On success will get the result whether the images are same or not

    Cons faced while using:


    • This process takes time and completely depends on Internet Speed (ie. In India even in Cities there is not…
    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  15. loop through face landmarks

    As of right now it seems like it is only possible with the SDK to call a landmark position as followed:
    face.FaceLandmarks.NoseTip.X

    to get all the locations I have to do this 54 times. Also in the future if something changes in this path, the application breaks. If I could loop through the FaceLandmark it could be way easier.

    For example:
    face.FaceLandmarks[0].X
    Or
    face.FaceLandmarks["NoseTip"].X

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

    We’ll send you updates on this idea

    0 comments  ·  Samples & SDK requests  ·  Flag idea as inappropriate…  ·  Admin →
  16. First time Face detection has high latency (5-6 secs)

    I am getting a detection latency of 5+ secs to detect the face in a photo when sending the request for the first time. As the image is stored temporarily for 24 hours, the subsequent detection requests for the same face is faster. However, the latency for the first request doesn't suffice our use case (or many real-world use cases in general).
    I am currently located in Dallas and I am using West Central endpoint. I have attached a screenshot of the time taken for the request for your reference. Please let me know if I am doing anything wrong…

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  17. Ponto de Extremidade

    "Esta chave de API está ativa no momento", mas quando eu adiciono o ponto de extremidade, não è encontrado a pagina? Alguem sabe como resolver??

    JSON Response

    {
    "error": {

    "code": "404",
    
    "message": "Resource not found"

    }
    }

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  18. Have a product you can find

    Have a product you can find

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

    We’ll send you updates on this idea

    0 comments  ·  Samples & SDK requests  ·  Flag idea as inappropriate…  ·  Admin →
  19. Support HEIC/HEVC format

    Please support the heic image format. Images taken trough iOS are standard in HEIC/HEVC format.

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

    We’ll send you updates on this idea

    0 comments  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
  20. not working Api for local files (base64 binary encoded data)

    Not working Api for local files (base64 binary encoded data)

    I converted test image from PHP example to base64 binarry data: (https://upload.wikimedia.org/wikipedia/commons/3/37/Dagestanimanand_woman.jpg) and after sending it to to endpoint I got always :

    "error": {

        "code": "InvalidImage",
    
    "message": "Decoding error, image format unsupported."
    }

    }

    When I use in url mode it's working well , but in our project we can't use this toll in that way. Please post working example for PHP with passing local binary data of file.

    --
    regards
    Peter

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

    We’ll send you updates on this idea

    1 comment  ·  API  ·  Flag idea as inappropriate…  ·  Admin →
← Previous 1 3 4 5 6
  • Don't see your idea?

Feedback and Knowledge Base