Text to Image Search | Coactive AI

curl -X POST https://api.coactive.ai/api/v1/search/text-to-image \
     -H "Authorization: Bearer <token>" \
     -H "Content-Type: application/json" \
     -d '{
  "dataset_id": "c40276f0-024b-4a3f-b3e6-dcf0d304843e",
  "text_query": "Dog"
}'

{
  "data": [
    {
      "created_user_id": "user_example",
      "created_dt": "2018-09-11T15:11:45.456777Z",
      "updated_user_id": "user_example",
      "updated_dt": "2018-09-11T15:11:45.456777Z",
      "coactive_image_id": "a6894d72-612d-4c20-b4d0-8f3f21c2bb10",
      "path": "s3://your-s3-bucket/path/to/video.mp4",
      "metadata": {
        "artist": "bufo frog"
      },
      "moderation_score": 0.1
    },
    {
      "created_user_id": "user_example",
      "created_dt": "2018-09-11T15:11:45.456777",
      "updated_user_id": "user_example",
      "updated_dt": "2018-09-11T15:11:45.456777",
      "coactive_image_id": "a6894d72-612d-4c20-b4d0-8f3f21c2bb10",
      "moderation_score": 0.1,
      "audio_segment": {
        "end_time_ms": 1000,
        "speech_to_text_transcription": "Winter is coming.",
        "start_time_ms": 0
      },
      "keyframe_is_v1": false,
      "shot": {
        "end_time_ms": 1000,
        "shot_id": "5007b9b7-0d74-4054-bbc7-cd9d2333ca8f",
        "start_time_ms": 0
      },
      "timestamp": 1029849493,
      "video": {
        "coactive_video_id": "5cf18348-8336-4db7-bb35-db7fbfe8951d",
        "created_dt": "2018-09-11T15:11:45.456777Z",
        "created_user_id": "user_example",
        "metadata": {
          "episode": 1,
          "season": 1,
          "show": "The Boys",
          "title": "The Name of the Game"
        },
        "path": "s3://your-s3-bucket/path/to/video.mp4",
        "updated_dt": "2018-09-11T15:11:45.456777Z",
        "updated_user_id": "user_example"
      }
    }
  ]
}

Search a dataset with a natural-language query and get back the images that best match it. Use this when you want to find visuals by describing what’s in them.For example, search for a sunset at the beach or Jalen Hurts holding the Vince Lombardi trophy. Pair the text query with exact-match metadata filters to narrow results further.

Authentication

AuthorizationBearer

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

Request

This endpoint expects an object.

dataset_idstringRequiredformat: "uuid"

The unique identifier for the dataset

text_querystringRequired

The text query to search for

offsetintegerOptional0-9223372036854776000Defaults to 0

Starting index to return

limitintegerOptional1-1000Defaults to 100

Max number of items to return

metadata_filtersobject or nullOptional

List of metadata filters to apply to the search

moderation_score_typeenumOptional

Type of moderation scores to return when moderation is enabled.probability returns scores 0-1 using BGE Reranker model.level returns scores 0-5 using OpenAI model.This does not enable/disable moderation - use skip_moderation for that.

negative_textstring or nullOptional

Optional text describing what to exclude from search results. Steers the query embedding away from this direction.

asset_typeenum or nullOptional

The asset type to filter results by. If no asset type is selected the endpoint will search over images and videos

skip_moderationbooleanOptionalDefaults to false

Skip content moderation if enabled

Response

Successful Response

datalist of objects

The paginated results

Errors

422

Unprocessable Entity Error