S9 Vision Labels

Detect labels and captions in image (Fast = Brainiall object detection module closed-set, Standard = Brainiall Vision Tagger engine multi-task)

Fast tier: Brainiall object detection module (COCO 80 classes, 236 ms). Standard tier: Brainiall Vision Tagger engine multi-task (caption / detect / OCR / segment / zero-shot text prompts, 1.6 s). Capability AWS Rekognition does not offer: zero-shot prompting + dense captions. permissively licensed.

post/v1/vision/labels/base64

Request

  • Base URL: https://api.brainiall.com
  • URL: https://api.brainiall.com/v1/vision/labels/base64
  • Auth: HTTP bearer

Request body

imagestring required

Base64-encoded image

tier'fast' | 'standard'

fast = Brainiall object detection module closed-set (COCO 80 classes) · standard = Brainiall Vision Tagger engine multi-task (caption / detect / OCR / segment / zero-shot)

taskstring nullable

Brainiall Vision Tagger engine task tag (Standard tier only). E.g. <CAPTION>, <DETAILED_CAPTION>, <OD>, <OCR>, <REFERRING_EXPRESSION_SEGMENTATION>

promptstring nullable

Optional zero-shot text prompt (Standard tier with referring expression tasks)

Response

Labels and/or caption returned

request_idstring
processing_msinteger
tier'fast' | 'standard'
captionstring nullable

Brainiall Vision Tagger engine caption (Standard tier)

outputobject nullable

Brainiall Vision Tagger engine raw multi-task output

Changes

No recorded changes to this operation across all 1 revision of this API.