Google Cloud Vision

Use Google Cloud Vision through a saved Integration hub connection. The catalog includes 8 actions, listed below.

What you can do

Use Google Cloud Vision through a saved Integration hub connection. The catalog includes 8 actions, listed below.

Documents and AI

Authentication: API key (enter the key and other details requested by the connection form)

Actions

Action	Description
Annotate file annotate_file	OCRs a PDF, TIFF or GIF file and returns the text of each page.
Annotate image annotate_image	Runs any combination of Vision features on one image (objects, landmarks, web matches, colors, labels, text).
Detect document text detect_document_text	Performs high-accuracy OCR optimized for documents such as invoices, forms, or contracts.
Detect faces detect_faces	Detects faces and emotional attributes in an image.
Detect labels detect_labels	Detects general objects and concepts (labels) in images, such as “car”, “tree”, or “building”.
Detect logos detect_logos	Detects brand logos in images, useful for brand monitoring or compliance automation.
Detect safe search detect_safe_search	Detects adult, violent, or explicit content in images for moderation and compliance workflows.
Detect text detect_text	Detects text within an image using Google Cloud Vision OCR.

How to connect

Open Integration hub, choose Google Cloud Vision, and create a named connection. Fill the key and any additional account details requested by the connection form. Save the connection.

For the shared connection process, follow Connect an app.

Use it in a workflow

Add an Integration step. Choose the app, saved connection and action, then map the action inputs from your record or workflow data.

The action list comes from the integration catalog. The connection form supplies the provider-specific input fields.