Generate a video description | Coactive AI

import requests
url = "https://api.coactive.ai/api/v0/video-narrative-metadata/datasets/dataset_id/videos/video_id/description"
payload = { "video_context": "This is a 170-minute film directed by Michael Mann in 1995, starring Al Pacino and Robert De Niro." }
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.json())

{
  "description": "The video shows armed robberies, police investigations, surveillance, and planning meetings across Los Angeles. Gunfire, sirens, and traffic noise dominate action scenes, while quieter moments take place in apartments, diners, and cars. The story ends with a nighttime chase at an airport runway, where Hanna confronts McCauley."
}

Generate a video description.

Description: A very detailed, factual account of what is visually, verbally, and audibly happening in the video. It accurately describes scenes, actions, environments, sounds, and dialogues without interpretation.

Authentication

AuthorizationBearer

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

Path parameters

dataset_idstringRequiredformat: "uuid"

video_idstringRequiredformat: "uuid"

Request

This endpoint expects an object.

video_contextstring or nullOptional

Optional context about the video

Response

Successful Response

descriptionstring

Generated video description

Authentication

Path parameters

Request

Response

Errors