Getting started with Adobe Smart Tags

Abhigyan Swaroop
3 min readMay 30, 2020

Adobe sensei offers a very powerful capability to auto-tag assets which helps the business users to quickly find the assets based on taxonomy.

In the recent years, Adobe has been making a lot of efforts to improve the assets' capability and make it more interactive for both customers and business authors. In this journey, a significant push is given by using artificial intelligence (AI) and machine learning (ML) framework which enables auto tagging of assets by applying both Standard and business specific tags to imagery.
This capability of adobe sensei called Smart Tags can also identify content, individual words, or phrases and can automatically apply descriptive tags to them.

Image courtesy : Adobe

Prerequisite

An Adobe ID account that has administrator privileges for the organization.

The Smart Content Service service is enabled for your organization.

Typical set-up

Basically, there are 3 main steps involved in enabling smart tags on your assets.
1. Enable Smart tag service
2. Train Smart tagging Algorithm
3. Add smart tags on your assets

Important things to keep in mind while performing above steps

1. Enable Smart tag service

a). Login to https://adminconsole.adobe.com/ via your organization id
After successful login, you should be redirected to an organization overview page. Verify the organization on top Right corner, to assure you are accessing the desired organization admin console.
b). In the overview link, look for “Smart Content Services”. If it appears, the service is enabled.

c). Follow link obtain public certificate
d). Follow link to create Adobe I/O integration
e).
Follow link configure smart content service
f).
Follow link to validate the configuration.

2. Train Smart tagging Algorithm

a). Use set of approx. 30 similar looking images(preferably object in focus) with different background.
b). Avoid using same tags on dissimilar looking images, it may lead to inaccurate tagging.
c). Add all different tags qualifying for the asset at once and avoid multiple training session on same image as it may affect the previous trained tags.
d). The training workflow runs on folders only.
e). Follow link to
generate report for trained tags.
f). Follow link for
Additional column can be added to report

3. Add smart tags on your assets

a). Enable smart tagging in the DAM Update Asset workflow
b). The tagging workflow runs on both assets and folders.

c). Open the properties page of your asset folder, select Enable Smart Tags under the Details tab, and save the changes.
d). In the subsequent tagging cycles, only the modified assets are tagged again with newly-trained tags. However, even unaltered assets are tagged if the gap between the last and current tagging cycles for the tagging workflow exceeds 24 hours. For periodic tagging workflows, unaltered assets are tagged when the gap exceeds 6 months.

Limitations of Smart Tagging

Enhanced smart tags are based on learning models of brand images and their tags. These models are not always perfect at identifying tags. The current version of the Smart Content Service has the following limitations:

1. Inability to recognize subtle differences in images. For example, slim versus regular fitted shirts.
2. Inability to identify tags based on tiny patterns/parts of an image. For example, logos on T-shirts.
3. Tagging is supported in the locales that AEM is supported in. For a list of languages, see
Smart Content Services release notes.

Official document:
Smart Tags

Stay tuned for more similar quick reads.

--

--