Using text recognition while also using object detection

Google Cloud Build with AI Custom ML & MLOps
1

Can you use visionAI text recognition while using object detection, or will they have to be two separate calls the Vertex AI?

2

Hi,

You will be needing calling text recognition and object detection separately. Endpoints for text recognition and object detection are separate hence you need to call them separately. See text recognition example and object detection example.

3

You can actually to multiple recognitions with one API call.
Both text recognition and object detection using the same API endpoint.

POST https://vision.googleapis.com/v1/images:annotate

The API endpoint request body has a parameter called features.
Which is an array and can hold multiple features.
see documentation
https://cloud.google.com/vision/docs/reference/rest/v1/AnnotateImageRequest

(The type of Google Cloud Vision API detection to perform, and the maximum number of results to return for that type. Multiple Feature objects can be specified in the features list.)

see documentation:
https://cloud.google.com/vision/docs/reference/rest/v1/Feature

As response you get all the features you requested
https://cloud.google.com/vision/docs/reference/rest/v1/AnnotateImageResponse

Enjoy
Sascha

1 Like