AI image tagging automation (video tutorial)

Overview

Learn how to use MediaFlows to automate AI Vision tagging in Cloudinary. This tutorial demonstrates how to set up a workflow that automatically analyzes uploaded images and applies relevant tags based on custom prompts, eliminating the need for manual image classification.

In this example, you'll see how a fashion or e-commerce company can automatically tag model images as "front-facing" or "back-facing" based on AI analysis. This automation ensures consistent tagging across your entire image library and enables powerful filtering and search capabilities.

Video tutorial

Tutorial contents

This tutorial presents the following topics. Click a timestamp to jump to that part of the video.

Introduction to AI image tagging

0:00	Discover how to automate AI Vision tagging using MediaFlows. This demo shows how to automatically analyze and tag images upon upload, eliminating manual classification work and ensuring consistent tagging across your asset library.

Configure the upload block

0:06

Configure the Cloudinary upload block to define when the MediaFlows automation should be triggered. The configuration allows you to specify exactly which uploads should activate the tagging workflow. You can trigger the automation on every upload, or use specific filters to target only certain asset types (images vs. videos), specific folders, or other criteria to ensure precision in your automation. In this case, set the upload block to trigger only for images being uploaded to a specific folder.

Configure the AI Vision block

1:07	The AI Vision Tag By Prompts block analyzes each uploaded image using custom prompts and applies tags to the asset. The block processes all assets that pass through the Cloudinary upload trigger you configured.

Create AI prompts

1:33	This example demonstrates two prompts: "Is the model front facing?" and "Is the model back facing?" The AI evaluates each image against these prompts and automatically applies tags based on the responses. See the AI Vision documentation for more prompt examples.

Test the automation

2:07	Test your automation by uploading sample images to the designated folder. The MediaFlows workflow processes each image in the background, analyzing them with AI Vision and applying the appropriate tags without any manual intervention.

View automated results

2:30	Review the automation results by refreshing the folder. Each image now has the correct tags automatically applied—"front-facing" or "back-facing"—based on the AI analysis. These tags enable powerful search and filtering capabilities for your team.

Understand the logs

2:59	Examine the MediaFlows logs to understand how the automation works. The logs show the output of each block in the workflow, including the responses from Cloudinary APIs and AI Vision analysis. This transparency helps you troubleshoot and optimize your automations.

Keep learning

If you like this, you might also like...

 

 

Check out the Cloudinary Academy for free self-paced Cloudinary courses on a variety of developer or DAM topics, or register for formal instructor-led courses, either virtual or on-site.