The technology no one else in VTON has.

The uncomfortable truth about VTON

Showing a customer how a garment looks on them does not tell them whether it will fit. The VTON industry tiptoes around this, but it is the truth. Without a size recommendation, the customer is still guessing, still getting it wrong, and still sending the garment back. Reverse logistics still bleeds your P&L.

Virtual try-on without fit intelligence is a beautiful demo with no business outcome.

This is the reason Tuck exists. We built the measurement engine before we built the visual layer. Try-on is the seduction. Fit Intelligence is the value.

What it actually does

From a single photograph, taken at a kiosk, on a phone, or uploaded online, Fit Intelligence reconstructs the customer's real body geometry, more than fifty points of it. From that, it generates a precise measurement profile covering chest, waist, hips, shoulder width, sleeve length, inseam, neck, bicep, thigh, calf, and more.

It then compares those measurements not to a generic chart but to your size chart, for that specific garment, in your specific cut and fabric. It visualises the fit on a body heat map: green for perfect, yellow for snug or relaxed, orange for tight or loose. And it recommends the size, with an explanation of why.

For maximum precision, an optional A-pose secondary capture unlocks even tighter measurement tolerances.

How we built it

Two models, sharing one underlying intelligence.

The standard model runs on a normal phone or kiosk camera. It is cost-efficient, production-ready, and is what most customers will deploy.

The high-accuracy model runs on advanced cameras and depth sensors and is for the cases where measurement tolerances need to be tight enough for premium or made-to-measure use cases.

Both have been tested across body types, lighting conditions, camera grades, postures, clothing colors, and the kind of everyday human error that breaks lab-grade systems the moment they meet the real world. They survive it. That is the entire point of the years we spent on this.

The fit heat map is the single most powerful trust-builder we ship, because it does not just give the customer a recommended size. It shows them why. Customers do not argue with diagrams the way they argue with size charts.

We also identify body type and skin tone, and use that to suggest styles and palettes that flatter the shopper. That drives item discovery and average order value.

Sizing is brand-aware. Plug your size chart in once, and the recommendations are tuned to your cuts, your fabrics, your tolerances, not to an "industry average" that has never quite fit anyone.

Customer measurements are processed for the session. We do not monetise them. There is no third-party data transfer.

Where it lives

Fit Intelligence is the engine running underneath everything else we make. It powers size recommendation on every product page where the Tuck VTON API is deployed. It powers the "See My Fit" heat map at every Magic Mirror kiosk. And it is available as a standalone API for brands that want to integrate it into their own apps, sizing widgets, or mobile experiences.

What it does to your P&L

Returns from "wrong size" drop 25 to 40 percent. The "buy 2, return 1" behaviour, which is currently treating your warehouse like a free fitting room, disappears. Customer trust shifts from guessing to confidence. Reverse logistics costs fall by roughly a third. Repeat purchase rates climb, because customers who bought the right size the first time tend to come back.

Why beta is the right time to join

We are rolling Fit Intelligence out with a small group of brand partners before general availability. Beta partners get priority onboarding and dedicated integration support. They get co-development input on size-chart calibration for their specific catalogs. They lock in founding-customer pricing for the long term. And they get a first-mover position in a category that, eighteen months from now, every fashion brand will have to deploy.

Who should be talking to us

Brands with return rates above 20 percent that are bleeding margin to size confusion. D2C labels where uncertainty about sizing kills first-time conversion. Retailers expanding into online private label. Apparel marketplaces that need a size-confidence layer across thousands of SKUs they don't fully control.