Skip to content
Video Editing
/
/
Generate Mask by Key Poin...

Generate Mask by Key Points

Request

Generates a segmentation mask video based on coordinate key points.

Response behavior:

  • Returns HTTP 202 with request_id and status_url
  • The final result (via status endpoint) will contain a mask_url.

Constraints:

  • Max Duration: 5 seconds (Truncated if auto_trim=true).
  • Max Resolution: 750p.
Security
api_token
Bodyapplication/jsonrequired
videostringrequired

Publicly accessible URL for the input video file.

frame_indexinteger

The frame number to apply key points to.

key_pointsArray of objectsrequired

An array of coordinate objects defining the mask hints.

  • positive: Inclusion points. These coordinates identify areas that SHOULD be part of the mask.
  • negative: Exclusion points. These coordinates identify areas that SHOULD NOT be part of the mask.
auto_trimboolean

If true, trims the video to the 5s limit.

Default:false
output_container_and_codecstring
Default:"mp4_h264"
Enum:"mp4_h264""mp4_h265""webm_vp9""mov_h265""mov_proresks""mkv_h264""mkv_h265""mkv_vp9""gif"
webhook_urlstring, (uri)

Optional URL for receiving the result via webhook when the async job completes. See webhooks.

POST
/mask_by_key_points
from bria_client import BriaSyncClient

client = BriaSyncClient()  # reads BRIA_API_TOKEN

job = client.submit(
    endpoint="video/segment/mask_by_key_points",
    payload={
        "video": "example",
        "frame_index": 88,
        "key_points": [
            {
                "x": 650,
                "y": 400,
                "type": "positive",
            },
            {
                "x": 655,
                "y": 405,
                "type": "positive",
            },
            {
                "x": 50,
                "y": 100,
                "type": "negative",
            },
        ],
    },
)
result = client.poll(job, interval=5, timeout=600)
print(result.result.video_url)

Responses

Accepted

Bodyapplication/json
request_idstringrequired
status_urlstringrequired
Response
{
  "request_id": "string",
  "status_url": "string"
}