The photograph is the
entire catalogue intake.
A shelf photograph enters the platform. Products come out as structured, ready-to-publish listings - name, unit, price, category, stock state. No forms. No spreadsheets. No typing.
What the intake actually recognises
A shopkeeper never has to teach the system what a jar of pickle looks like. The recognition operates at the level a shopkeeper thinks - the product, not the pixel.
Common consumer brands, staples, personal care, dairy, snacks and household - identified with their canonical name, brand and pack size.
Grams, millilitres, count, pieces per pack - inferred from the pack itself, not asked for separately.
MRP and shop-price are attached from your last confirmation. New products inherit a suggested price you can override in a tap.
Listings are grouped into the categories buyers actually search - Staples, Dairy, Bakery, Personal Care - not internal taxonomies.
Stock state, freshness where applicable, and low-stock flags are inferred from the shelf and confirmed at review.
A single frame of a mid-shelf in a Pune kirana shop. Six items visible, none typed. This is what the shopkeeper sees on the review screen a moment later.
A shelf image is uploaded from a phone. No cropping or straightening is required.
The system identifies six named products, assigns units and prices, and drops them into categories.
The shopkeeper reviews, adjusts one price, taps publish. The storefront now carries all six.
Tell us what kind of shop you run.
Watch the storefront appear.
Pick a shop type. We rebuild a shelf, generate listings, and draw a realistic delivery radius - all in the same moment. This is exactly how your own store comes online.