Face recognition - find_similar operation on PersistedFaceId

Question

For face recognition task I need to use find_similar operation on 2 collections.

I would like to merge them before searching the face.

I merged the PersistedFaceIds of the two collections into one list and them transfer it to "find_similar" method but got and error.

Does "find_similar" method can work on list of PersistedFaceIds? If not , there is a workaround to search in 2 collections in one call?

Answer

this one works well-

        matched_faces = face_client.face.find_similar(

            face_id=face_id_source,

            face_list_id=collection_id,

            max_num_of_candidates_returned=5,

            recognition_model="recognition_04",

            detection_model="detection_03",

        )

But if I use this one

   # collection_face_ids is a list of all persisted_face_id in collection_id

     matched_faces = face_client.face.find_similar(
            face_id=face_id_source,
            face_ids=collection_face_ids,
            max_num_of_candidates_returned=5,
            recognition_model="recognition_04",
            detection_model="detection_03",
        )

I get following error

  File "C:\Anaconda3\envs\face_recognition\lib\site-packages\azure\cognitiveservices\vision\face\operations\_face_operations.py", line 136, in find_similar
    raise models.APIErrorException(self._deserialize, response)
azure.cognitiveservices.vision.face.models._models_py3.APIErrorException: (FaceNotFound) Face is not found.

What I want to do is to search face_id_source in collection_id1and collection_id2 in one call so I tried to do

combined_face_ids = collection1_face_ids + collection2_face_ids

matched_faces = face_client.face.find_similar(
face_ids=combined_face_ids,   
 max_num_of_candidates_returned=5,    
recognition_model="recognition_04",    
detection_model="detection_03",)

and get the above error

Share via

Face recognition - find_similar operation on PersistedFaceId

1 answer

Your answer