Skip to main content

Label Recognition

Thomas Goblet avatar
Written by Thomas Goblet
Updated this week

Label Recognition is a handy feature that helps your team capture and manage key product details straight from garment labels, automatically. Powered by AI, it reads information like material composition, care instructions, country of origin, and more, then turns it into digital data. It cuts down on manual work and reduces errors, giving your team more time to focus on what matters.

Just a heads-up: Label Recognition is an add-on. If you’re interested, reach out to your Customer Success Manager for more info.


Current Capabilities

Label Recognition currently supports:

  • Fashion products

  • Garment labels

  • Identifying Care Instructions, Material Composition, Country of Origin, Manufacturer, Warnings and Package Contents

Extracting Label Information With Label Recognition

Access via the Chroma app

You can get started with Label Recognition right in the Chroma application.

After scanning a sample’s barcode on the Scan page, you’ll see a “Detect Label” option. This might show up right below the scan...


...or inside the expanded list.


You can either snap photos of the label on the spot or upload them from your gallery. Each sample can have up to 10 images.


As soon as you hit upload, the system starts analyzing them automatically.


If the images are blurry or hard to read, Chroma will let you know that it couldn't extract the data.


After the upload, you’ll still be able to open and view the uploaded images in full screen by tapping on “View”.

Everything uploaded syncs automatically with Gamma in the background.


Access via Gamma

Initiating Label Recognition as part of a Scan flow

If you’re working in Gamma, you can also add label images during a regular Sample scan, where you can now add an additional step to upload the Label images.


When you get to the Product Label step, Gamma will generate a QR code. Just scan it with Chroma, and it’ll open the Detect Label flow, as seen above, for that Sample.

Upload your images, and the system will handle the rest, syncing and analyzing as it goes.


Initiating Label Recognition from the Sample Slide-In Panel

You can also access Label Recognition directly from a sample’s page in Gamma.

When you click on a Sample and land on the Overview tab, scroll to the Label Recognition section and hit “Add.”


From there, you can either upload images from your computer or scan the QR code with Chroma to upload from your phone.


Once the images are detected, Creative Force will get to work analyzing them.


Reviewing and Editing Extracted Label Data

After your images have been uploaded, whether through Chroma or directly in Gamma, you’ll see them appear in the Overview tab of the Sample slide-in.

The system will begin analyzing the images right away, and you’ll see that status reflected in the interface.

When the analysis is complete, the results will be visible in the same panel.


If any of the images are too blurry or unclear, Creative Force will display a “No Data” message to let you know it couldn’t extract information.

You can also update the photos themselves if you'd like to submit clearer images.


If the system does manage to extract data, it’ll appear directly next to the images.

Here you can review the details, and if needed, update or correct them.


Clicking “Edit Data” will let you select which fields to adjust and overwrite anything the tool may have misread.



Auto-Filling Detected Label Values Into Copywriting Fields

If a Product is linked to a Sample that has recognized label data, and that product also has a copywriting task (either in “To do” or “Backlog” status), the system will prefill fields like Material Composition and Care Instructions automatically.


There is a small caveat: the values from the label need to match what’s set up in your Copywriting fields. If they don’t match exactly, the system will highlight the field with an orange warning for you to review.


You can also use merge field tokens to pull in label data into your copy. These can fill in single-line or text area fields automatically, or they can be added manually while writing.

Did this answer your question?