Skip to main content
POST
/
face-detect
Detect faces in video
curl --request POST \
  --url https://api.cloudglue.dev/v1/face-detect \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "url": "<string>",
  "frame_extraction_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "frame_extraction_config": {
    "strategy": "uniform",
    "uniform_config": {
      "frames_per_second": 1,
      "max_width": 1024
    },
    "thumbnails_config": {
      "enable_frame_thumbnails": true
    },
    "start_time_seconds": 1,
    "end_time_seconds": 1
  }
}'
{
  "face_detection_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "frame_extraction_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "file_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "status": "pending",
  "created_at": 123,
  "data": {
    "object": "list",
    "total": 123,
    "limit": 50,
    "offset": 1,
    "faces": [
      {
        "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
        "face_bounding_box": {
          "height": 0.5,
          "width": 0.5,
          "top": 0.5,
          "left": 0.5
        },
        "frame_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
        "timestamp": 123,
        "thumbnail_url": "<string>"
      }
    ]
  }
}

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Face detection request parameters

url
string
required

URL of the target video to analyze

frame_extraction_id
string<uuid>

Optional: Use previously extracted frames

frame_extraction_config
object

Optional: Frame extraction configuration

Response

Face detection job created successfully

face_detection_id
string<uuid>
required

Unique identifier for the face detection job

status
enum<string>
required

Status of the face detection job

Available options:
pending,
processing,
completed,
failed
created_at
number
required

Unix timestamp of when the job was created

frame_extraction_id
string<uuid>

ID of the frame extraction used

file_id
string<uuid>

ID of the target file

data
object

Face detection results (only present when status is completed)

I