Floor Plan Recognition

Mapping a store by hand takes time. Floor Plan Recognition lets you upload or photograph a store directory map, and AI automatically identifies aisles, entrances, checkouts, and the store boundary -- giving you a head start on building an accurate store layout.

How It Works

Take a photo of the store's directory sign (the kind you see near the entrance) or upload an image of a floor plan. The AI vision system analyzes the image and produces a set of recognized features in the correct positions. You get a preview overlaid on the map where you can review, adjust, and rename features before committing them to your store layout.

Using Floor Plan Recognition

1. Open the store editor -- Navigate to Map Your Store and open the store you want to map (or create a new one).
2. Click "Scan Floor Plan" -- Find this option in the drawing tools panel.
3. Upload or capture an image -- Choose a photo from your device or take one on the spot. Supported formats include JPEG, PNG, GIF, and WebP.
4. Wait for analysis -- The AI processes the image and identifies store features. This usually takes a few seconds.
5. Preview the results -- See the recognized features overlaid on the map. Each feature is labeled with a name and category.
6. Adjust as needed -- Rename aisles, change categories, or remove features that were incorrectly identified.
7. Apply to your store -- When you're satisfied, apply the recognized features to your store layout.

What Gets Recognized

The AI looks for common store layout elements:

• Aisles -- Identified with their approximate position and labeled with a likely category (produce, dairy, bakery, etc.).
• Entrances -- Store entry points.
• Checkouts -- Checkout lane areas.
• Store boundary -- The overall outline of the store.

Boundary Hints

If you've already drawn a store boundary on the map, Floor Plan Recognition uses it as a reference to position the recognized features accurately within your existing layout. This is especially useful when the photo doesn't capture the exact scale of the store.

Tips

• Clear, well-lit photos of store directories work best. Avoid extreme angles -- a straight-on shot gives the most accurate results.
• The AI does its best with typical grocery store layouts. Unusual floor plans or heavily stylized directory designs may need more manual adjustment.
• Use this feature as a starting point, not a finished product. Walk the store to verify aisle positions and categories after applying the recognized layout.
• If the store has a multi-floor layout, scan one floor at a time for best results.
• Photos of printed store directories tend to work better than hand-drawn maps, but both are supported.