Function

Create a video celebrity analysis job to recognize and analyze political figures and celebrities in videos and warn you of sensitive content in the videos. This helps you quickly locate and handle sensitive content in videos and reduce any non-compliance risks in your business.

• Supported video formats: AVI, WMV, MPG, MPEG, MP4, MOV, M4V, MKV
• Videos cannot be stored in OBS buckets encrypted by KMS.
• The size of a single video cannot exceed 4 GB.

• When data is read from a specified URL, the video size cannot exceed 1 GB.
• The facial image must be at least 40 x 40 pixels.
• Faces can be recognized with up to ±15° pitch and ±30° yaw.
• A maximum of 20 faces can be identified at the same time.
• Supported regions: CN North-Beijing1 and CN North-Beijing4.

URI

• URI format:

  POST /v2/{project_id}/services/video-celebrity-recognition/tasks

• Parameters

  Parameter	Mandatory	Type	Description
  project_id	Yes	String	Project ID corresponding to the region where the service is located. For details about how to obtain the project ID, see Obtaining the Project ID.

Request Parameters

• Example request

  POST /v2/6204a5bd270343b5885144cf9c8c158d/services/video-celebrity-recognition/tasks  
   {  
       "name": "task-est",  
       "description": "description",  
       "input": {  
              "type": "obs",  
              "data": [  
                     {  
                      "bucket": "obs-iva",  
                      "path": "input/demo.mp4"  
                     }  
              ]  
       },  
       "output": {  
              "obs": {  
                     "bucket": "obs-iva",  
                     "path": "output/"  
              }  
       },  
      "service_version":"1.0"  
   }

• Parameters

  Name	Mandatory	Type	Description
  name	Yes	String	The task name. It can be up to 100 characters long and can include Chinese characters, letters, digits, underscores (_), and hyphens (-).
  description	No	String	A task description. It can be up to 500 characters long.
  input	Yes	Object	List of input video data. Currently, the following input types are supported: obs: Read video data from OBS on HUAWEI CLOUD. The video size cannot exceed 4 GB. Videos encrypted by KMS cannot be stored in OBS buckets. url: Read video data from a specified URL. The video size cannot exceed 1 GB. Currently, only the URL of OBS is supported. You need to grant the anonymous users the permission to read the URL. For details, see Bucket ACL Overview. For details, see task.input.
  service_version	Yes	String	Function version: 1.0.
  output	Yes	Object	List of output result data. Currently, only the following output types are supported: obs: Export the results to the OBS bucket you specified. hosting: Host the results on OBS at the service side. The OBS path is specified by the service. You can obtain the path by using the API that queries a single task. For details, see Querying a Single Job. For details, see task.output.

Response Message

• Example response

  [    
     {      "id": "f18320e61e4c4dc685aa2dfc22a28dc5"   }  
  ]

• Parameters

  Name	Type	Description
  id	String	Task ID.

Video Analysis Results

• The video analysis results are saved, in JSON format, to a location you have specified. Typical file content is as follows:

  "frames": [
          {
              "url": "obs-dushubao/wxh_hjx/videopolitics/est/shijiuda30min.mp4.capture/3018.jpg",
              "offset": "00:02:00",
              "result": [
                  {
  "Label": "Xi Jinping",
                      "confidence": 0.6001852750778198,
                      "face_detail": {
                          "h": 71,
                          "w": 50,
                          "x": 302,
                          "y": 167
                      }
                  },
                  {
  "Label": "Jiang Zemin",
                      "confidence": 0.7005537152290344,
                      "face_detail": {
                          "h": 61,
                          "w": 48,
                          "x": 420,
                          "y": 247
                      }
                  }
              ]
          },
  ]

• Fields in the results file

  Field	Description
  frames	Video frame identification results
  url	Video frame storage location
  offset	Time when the identification was performed
  result	Information about the detected celebrity
  label	The name of the detected celebrity
  confidence	Results confidence
  face_detail	Facial details (in pixels) h: Area height. w: Area width. x: X-coordinate of the upper left corner of the area. y: Y-coordinate of the upper left corner of the area.

Status Codes

• Normal

  201

• Abnormal

  Status Code	Description
  400 Bad Request	Request error. For details about the returned error code, see Error Codes.
  401 Unauthorized	Authentication failed.
  403 Forbidden	No operation permission.
  404 Not Found	The requested resource was not found.
  500 Internal Server Error	Internal service error.
  503 Service Unavailable	Service unavailable.