Reka AI Documentation

import requests
url = "https://vision-agent.api.reka.ai/v1/images/search"
payload = { "query": "string" }
headers = {
    "X-Api-Key": "<apiKey>",
    "Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.json())

{
  "results": [
    {
      "image_id": "string",
      "image_url": "string",
      "image_metadata": {
        "camera_id": "string",
        "recorded_timestamp_utc": 1,
        "video_clip_id": "string",
        "org_id": "string",
        "site_id": "string",
        "event_id": "string",
        "caption": "string",
        "detection_metadata": [
          {
            "class_name": "string",
            "bounding_box": [
              1
            ],
            "confidence": 1.1,
            "is_license_plate": false,
            "license_plate_number": "string",
            "ncic_category": "string"
          }
        ]
      },
      "upload_timestamp": 1.1,
      "score": 1.1
    }
  ]
}

Search for images with a plaintext query

Authentication

X-Api-Keystring

API key authentication via X-Api-Key header

AuthorizationBearer

Bearer token authentication via Authorization header

Request

This endpoint expects an object.

querystringRequired>=1 character

Search query text

max_resultsintegerOptionalDefaults to 10

search_modeenumOptionalDefaults to vision

Search mode: ‘vision’ for vision-only search, ‘joined’ for hybrid search

Allowed values:

image_weightdouble or nullOptional

Weight of the image in the search (only used when search_mode=‘joined’)

text_weightdouble or nullOptional

Weight of the text in the search (only used when search_mode=‘joined’)

thresholddouble or nullOptional

Similarity score threshold

Response

Successful Response

resultslist of objects