Resources · Make integration

Amazon Rekognition Make integrationAutomate Amazon Rekognition with Make.

What if every image your team receives came back already tagged and read? The Amazon Rekognition Make integration gives you 5 modules: 3 actions and 2 searches that label images, pull text from pictures and videos, and flag unwanted video content. This guide connects your AWS account and walks you through a first scenario.

Try the Amazon Rekognition integration on MakeSee the 5 modules

Verified Trustpilot reviews · AI, automation & growth agency

What you can automate

What can the Amazon Rekognition Make integration do for you?

The Amazon Rekognition Make integration lets a Make scenario, the automation you build in Make as a chain of bricks called modules, hand an image or a stored video to Rekognition and get back what it found: objects, printed text or content that breaks your rules. You pick the source app, Make passes the file along, and the answer lands wherever you need it.

Searchable photo libraries. A property manager receives inspection photos from the field. With Detect labels in an image, each picture comes back with the things Rekognition recognized, so a later module can file it under the right tag instead of someone opening every shot.

Text pulled out of pictures. An accounts clerk gets snapshots of shop signs, labels or delivery slips. Detect text in an image turns the printed words into text a spreadsheet or CRM can store. For recorded clips, Detect text in a video does the same job on a stored video.

A first filter on user videos. A community platform accepts uploads from members. Moderate content in a video looks for inappropriate or offensive material, and your scenario can route anything flagged to a human reviewer before it goes live.

One thing to know before you start: Rekognition has no trigger in Make, so it never starts a scenario by itself. Another app, or the schedule you set on the first module, does that job. When the listed modules fall short, Make an API call reaches the rest of the Rekognition API with the same connection. n8n often offers a node for the same app, and our n8n vs Make comparison lays out the criteria. New to the editor? The Make training starts from zero, and the Make pricing page explains how operations are counted.

Connect

How do you connect Amazon Rekognition to Make?

  1. 01

    Create an access key in AWS

    Open the AWS Management Console, go to IAM, click Users and pick the user Make should act as. On the Security credentials tab, scroll to Access keys, click Create access key, then copy both keys shown in the window.

  2. 02

    Add a Rekognition module

    In your scenario, add any Amazon Rekognition module and click Create a connection. A connection is your AWS account linked to Make once, then reused by every Rekognition module you add later. The Connection name field is optional.

  3. 03

    Paste the credentials

    Enter the API key and the secret key you copied from AWS, or follow the instructions Make shows on screen, then save. The module can now send files to Rekognition on behalf of that AWS user.

First scenario

Your first scenario with Amazon Rekognition

GoalWhen a new image reaches the app where your team stores files, Make sends it to Amazon Rekognition and passes the detected labels to the next module.

  1. 01

    Start a new scenario

    From the Scenarios page, create a new scenario and click the +. Search for the app where your images arrive and pick its module that watches new files.

  2. 02

    Add Detect labels in an image

    Click the + on the right of that first module, search Amazon Rekognition and choose Detect labels in an image. Select the connection you created, or click Create a connection.

  3. 03

    Map the image

    Fill the module by mapping the file from the previous module, so each image that arrives is the one Rekognition analyzes. A bundle, one item passed from module to module, carries each file.

  4. 04

    Test with Run once

    Click Run once and drop a test image where the first module watches. Open the bubble on each module to read the bundles it received and check the labels Rekognition sent back.

  5. 05

    Schedule and switch on

    Set the schedule on the first module, every 15 minutes by default, then switch the scenario on. Each module run counts as one operation on your Make plan.

Modules

What each Amazon Rekognition module does

Amazon Rekognition gives you 5 modules. For each one: what it does for you, when to reach for it, and what to watch out for.

Amazon Rekognition1

Detect labels in an image

Action

In Make“Detects instances of real-world entities within an image provided as input.”

Give it an image and it returns the real-world things Rekognition spots in it, ready for the next module to use as tags or as a filter condition.

When to use it
a folder of product or site photos that nobody has time to tag by hand before they get archived.
Watch out
it reads objects, not words, so pick the text module when the picture holds printed information.
Amazon Rekognition2

Detect text in an image

Action

In Make“Detects text in the input image and converts it into machine-readable text.”

Turns the words printed in a picture into machine-readable text that your scenario can store, search or send to another app, instead of someone retyping a label or a sign.

When to use it
photos of delivery slips, shelf labels or badges that should end up as text in a sheet or a CRM record.
Watch out
run a test with Run once on a real photo first and read the bundle before you map the result further.
Amazon Rekognition3

Make an API call

Action

In Make“Performs an arbitrary authorized API call.”

Reaches any endpoint of the Amazon Rekognition API with the connection you already set up, which covers the features the four other modules leave out without building a separate authentication.

When to use it
you need a Rekognition operation that has no dedicated module in Make.
Watch out
nothing is prefilled here, so you describe the request yourself using the AWS documentation, and each call still counts as one operation.
Amazon Rekognition4

Detect text in a video

Search

In Make“Detects text in a stored video.”

Reads the text that appears in a stored video and hands it back to your scenario, so on-screen captions or titles become data you can index or check.

When to use it
a library of recorded clips where you want to find the ones showing a given word or product name.
Watch out
this is a search module, so it can return zero, one or several bundles, and the next modules run once per bundle.
Amazon Rekognition5

Moderate content in a video

Search

In Make“Identifies inappropriate, unwanted, or offensive content in stored videos.”

Scans a stored video for inappropriate, unwanted or offensive content and tells your scenario what it found, which lets you hold a clip back before anyone sees it published.

When to use it
members or customers upload videos and a person should review anything suspicious first.
Watch out
as a search, it can return zero bundles as well as several, so plan the next step for a clip with nothing flagged too.
Need help

Need help automating Amazon Rekognition with Make?

A person reads every message.

FAQ

Amazon Rekognition and Make: common questions

01Is the Amazon Rekognition Make integration free?
Yes, on the Make side. Amazon Rekognition is a standard app, so it is available from the Free plan, which allows 2 active scenarios, a minimum interval of 15 minutes between scheduled runs, 5 minutes of execution per run and files up to 5 MB. Paid plans lower the interval to 1 minute and remove the cap on active scenarios. Rekognition itself is billed by AWS under your own account, so check the Amazon Rekognition pricing page before you send large volumes.
02What does connecting Amazon Rekognition to Make require?
An active AWS account with Rekognition set up, and a user whose credentials have the right permissions. In the IAM section of the AWS Management Console, open that user, go to the Security credentials tab and create an access key. Copy both the API key and the secret key, then paste them in Make after clicking Create a connection on any Rekognition module. The connection name is optional and helps when you manage several AWS accounts.
03Does Amazon Rekognition have a real-time trigger in Make?
No. Amazon Rekognition has no trigger module in Make, scheduled or instant, so it cannot start a scenario by itself. The scenario begins with another app, for example the storage tool where images or videos land, or with the schedule on the first module, every 15 minutes by default. The Rekognition modules then run as steps in the middle of the flow, each one counted as an operation on your plan.
04What if a needed Amazon Rekognition module is missing in Make?
Use Make an API call. It reaches any endpoint of the Amazon Rekognition API with the connection you already have, so an operation without a dedicated module is still within reach. You write the request yourself from the AWS documentation. The Make documentation for this app also notes that part of its content is AI-generated, so verify the important details directly with Amazon Rekognition when something behaves differently than expected.
05Make or n8n for Amazon Rekognition?
It depends on where the rest of your automations already live. In Make, Rekognition is a standard app available from the Free plan, alongside 3,000+ standard apps, with 5 ready modules and Make an API call for the rest. n8n often has a node for the same app too. Compare the modules each tool offers for your exact case and pick the one your team already knows, rather than splitting the same process across two tools.
Same platform

Other docs you might need on this stack.

Browse all integrations