Vionlabs Docs
API Specs

Vionlabs Scene Tags API

External docs:SwaggerReDoc

Get item scene tags result status and data

GET
/results/scene_tags/v1/{inventory_id}

Gets scene tags (mood_tag, environment and keyword) aggregated over a time interval of a given title.

The pipeline computes a set of keywords per fixed (~30s) segment; this endpoint aligns those segments to the requested [start, end] interval, weighting each keyword by how much of the interval is covered by segments that contain it, and returns the top top_n tags per category ordered by descending weight.

From product version 2 the spec also includes nudity, blood-gore, profanity, and metadata-v1. Moderation flags are yes/no for the interval. Nudity and gore PTS keys are converted with the video stream time_base.

Authorization

AuthorizationBearer <token>

Stytch access token authentication. Requires BOTH:

  • Authorization: Bearer ACCESS_TOKEN
  • catalog query parameter

Example: Authorization: Bearer ACCESS_TOKEN ?catalog=my_catalog

In: header

Path Parameters

inventory_id*Inventory Id

Customer's identifier for a catalog item.

Query Parameters

start*Start

Start time of the interval to get scene tags for (seconds).

Range0 <= value
end*End

End time of the interval to get scene tags for (seconds).

Range0 < value
top_n?Top N

Maximum number of tags to return per category.

Default10
Range1 <= value <= 100
version?|

Specify scene tags model version code.

Response Body

application/json

application/json

curl -X GET "https://example.com/results/scene_tags/v1/string?start=0&end=1"
{
  "status": "success",
  "error": "string",
  "updated": "2019-08-24T14:15:22Z",
  "version_id": "string",
  "inventory_id": "string",
  "type": "movie",
  "data": {
    "inventory_id": "string",
    "type": "movie",
    "vionlabs_id": "string",
    "version_id": "string",
    "start_sec": 0,
    "end_sec": 0,
    "mood_tag": [
      {
        "tag": "string",
        "weight": 0
      }
    ],
    "environment": [
      {
        "tag": "string",
        "weight": 0
      }
    ],
    "keyword": [
      {
        "tag": "string",
        "weight": 0
      }
    ],
    "contains_nudity": true,
    "contains_gore": true,
    "contains_profanity": true
  }
}
Empty
Empty
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Get item scene tags result status

GET
/results/scene_tags/v1/status/{inventory_id}

Gets a scene tags result status (readiness) for a given asset.

Authorization

AuthorizationBearer <token>

Stytch access token authentication. Requires BOTH:

  • Authorization: Bearer ACCESS_TOKEN
  • catalog query parameter

Example: Authorization: Bearer ACCESS_TOKEN ?catalog=my_catalog

In: header

Path Parameters

inventory_id*Inventory Id

Customer's identifier for a catalog item.

Query Parameters

version?|

Specify scene tags model version code.

Response Body

application/json

application/json

curl -X GET "https://example.com/results/scene_tags/v1/status/string"
{
  "status": "success",
  "error": "string",
  "updated": "2019-08-24T14:15:22Z",
  "version_id": "string",
  "inventory_id": "string",
  "type": "movie"
}
Empty
Empty
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Get paginated list of items with scene tags result statuses

GET
/results/scene_tags/v1/

Gets a paginated list of items with scene tags result statuses for the customer catalog. Results are ordered by processing time of an asset, with the most recently updated assets first.

Authorization

AuthorizationBearer <token>

Stytch access token authentication. Requires BOTH:

  • Authorization: Bearer ACCESS_TOKEN
  • catalog query parameter

Example: Authorization: Bearer ACCESS_TOKEN ?catalog=my_catalog

In: header

Query Parameters

from_utcdatetime?|

Specifies start of time window (ISO 8601). If not specified then all results will be returned

to_utcdatetime?|

Specifies end of time window (ISO 8601). If not specified then utcnow() value will be used.It is recommended to fix this argument while looping across pages to establish a fixed time window

page_num?Page Num

Specifies a positional number of page with results

Default0
Range0 <= value
page_size?Page Size

Specifies a max number of results per page

Default500
Range1 <= value
include_episodes?|
Defaultfalse
version?|

Specifies scene tags version code

Response Body

application/json

application/json

curl -X GET "https://example.com/results/scene_tags/v1/"
{
  "page_num": 0,
  "page_size": 0,
  "total_count": 0,
  "data": [
    {
      "status": "success",
      "error": "string",
      "updated": "2019-08-24T14:15:22Z",
      "version_id": "string",
      "inventory_id": "string",
      "type": "movie"
    }
  ]
}
Empty
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Vionlabs Content Summary API

Vionlabs Content Summary API for summarizing information-heavy content (news segments, podcasts).

Vionlabs Vertical Conversion API

Vionlabs API for retrieving vertical conversion data, used to reframe a horizontal source video into a vertical crop without losing the subject of interest. For each frame, the response gives an x_center coordinate (in source-pixel units) saying where the camera should focus horizontally, plus the detected upper and lower black bar positions and the source frame dimensions. The API does not return a cropped video. Aspect ratio is independent of the conversion: the output is not tied to any specific output ratio (9:16, 4:5, 1:1, ...); the consumer picks one at render time by choosing how much of the source to include around x_center. Accordingly, the endpoint accepts no aspect-ratio, framing, or cropping parameters, only an optional version.