Recognition Settings

Configure which attributes AI recognition should extract, so the results go straight into your items' attribute values.

What Are Recognition Settings?

When you recognize an item you first pick a "category", and that category decides what information the AI should extract. Recognition Settings covers:

  • The master switch for recognition
  • Which attributes each category should extract
  • Preset Categories and the categories you create yourself
In Simple Terms

For example, AI recognizes "this is a bottle of shampoo". If you pick the "Personal Care" category, it extracts the attributes configured for that category (brand, volume, purchase channel...) and fills them into the item's attribute values. Everything recognition produces ends up as attribute values on the item — including category and location.

Accessing Recognition Settings

  • Tap the top-left corner of the home page to open the side menu
  • In the Settings group, tap "Recognition Settings"
  • Enter the configuration page

The top-right corner opens "Config Plaza", where you can browse and take categories shared by others; at the bottom of the page there is also "Clear all recognition results".

Screenshot: Recognition Settings page (enable switch, Preset Categories, Custom Categories)

Configuration Options

Enable Recognition

  • The "Enable recognition" switch at the top — recognition features work only when it is on
  • Non-members see a "Membership required" note under the switch

Preset Categories

  • A set of built-in categories that you can use right away
  • Open a category and the only thing to configure is which attributes it extracts
  • Add or remove attributes so the extracted information matches what you want to record

Custom Categories

  • No preset fits? Tap the floating button to create one
  • Name the category, then pick the attributes to extract (attribute name, type, options)
  • Select it when recognizing, and the AI works through that attribute list

Categories by Item Type

Create one category per kind of item and configure what each should extract:

  • Food: extract expiration date, brand, net content — expiration is what matters most
  • Clothing: extract color, size, material
  • Electronics: extract brand, model, warranty end date

Use Cases

Pick the right category for batch recognition

  • When AI image segmentation recognizes several items at once
  • Choose a category (preset or custom) that fits this batch first
  • Its attributes are extracted one by one, so there is much less to fill in afterwards

Auto-fill from barcode scanning

  • After scanning a barcode to retrieve product info
  • The information is written into the attributes configured for the chosen category
  • Speeds up item entry

Consistent data standards

  • Items of the same kind go through the same category, so the extracted attributes match
  • Entry format stays uniform naturally
  • Makes filtering and statistics easier down the line

Default Configuration

The built-in Preset Categories let beginners start right away:

  • Pick the preset category closest to the item type when recognizing
  • Its configured attributes are extracted and filled automatically
  • You can adjust those attribute values after recognition
Tip

Start with the preset categories, then fine-tune which attributes are extracted or build your own categories once you are comfortable. A well-configured attribute list means recognition results need no second round of cleanup.