Fashion · Single image · 5 credits

See the shoe on your feet, before you buy

User photo + shoe image. Stance preserved; shoe rendered at correct scale on the shopper's actual feet.

Try a virtual fitting See pricing

Stand shot in. Try-on out.

Every result below was generated from the photo next to it. No cherry-picking.

Input

Pixfino output

Input

Pixfino output

Input

Pixfino output

Input

Pixfino output

Highest-return-rate category

Footwear leads e-commerce returns. Try-on closes the visual-confidence gap most directly here.

Stance preserved

The customer's stance and lower body stay; only the footwear is replaced.

No AR SDK

Server-side render — works on any web PDP without mobile-app integration.

Who this is for

Sneaker brands

Job

Reduce stance-and-style returns

Friction

Shoes have the highest return rates in e-commerce.

Formal-shoe retailers

Job

Show how the shoe sits with the shopper's outfit

Friction

Customers can't tell suit-shoe coordination from product shots.

Athletic footwear

Job

Match the shoe to the customer's lifestyle photos

Friction

Performance footwear is bought by feel and look together.

Marketplaces

Job

Differentiate on per-listing try-on

Friction

AR SDKs require per-app integration.

How it works

1. 1

   Customer uploads opt-in photo

   Full-length or lower-body shot. Consent-aware.

   Replaces: Trying shoes in a physical store.

2. 2

   Pick the shoe

   From your catalog. Shoe_type (sneaker, formal, athletic, heel) parameters fit physics.

   Replaces: Carrying multiple sizes home.

3. 3

   Generate

   Around 30 seconds. Stance preserved; the shoe replaces existing footwear.

   Replaces: Returning a shoe that didn't suit the outfit.

4. 4

   Show on PDP

   Embed the try-on render on the PDP. Customer sees the shoe with their outfit.

   Replaces: Generic on-stand PDP imagery.

Pixfino vs traditional production

	Traditional studio	Stock + retouch	Pixfino
Lead time	Visit store	N/A	Under 1 minute
Per-fitting cost	Lost margin from returns	N/A	5 credits
Engineering effort	AR SDK	N/A	REST endpoint
Returns reduction	Baseline (high)	N/A	Measurable lift
Scale ceiling	Per-store throughput	N/A	Real-time
Customer effort	Visit, fit, return	N/A	One photo

Questions and answers

What photo does the user need to upload?

Full-length for dresses-with-heels coordination, lower-body shot for footwear-only try-on. Both are supported.

How accurate is the rendered shoe size?

The renderer scales the shoe to the user's foot in the photo. Style fit (does it suit me) is what try-on solves; physical sizing still requires the shopper to know their size.

What about consent?

Opt-in; the embed surfaces consent language.

Will the user's pants/dress stay correct?

Yes — only the footwear is replaced. The rest of the lower body is preserved.

Does this work for heels?

Yes — heel height affects ankle and calf positioning; the renderer simulates that.

What's the API surface?

POST /api/v1/presets/virtual-try-on-footwear/run with the user_photo_url and shoe_image.

Related templates

Catalog Variant Generator

Fashion · 3 credits

Diverse Model Variants

Fashion · 3 credits

Living Garment Video

Fashion · 64 credits

On-Model Product Shot

Fashion · 5 credits

Measured quality

Full QC report →

Virtual Try-On (Footwear)

oaktree/image-edit · $0.04/image

hard preset

On-model shoe try-on is hard: the render must keep the model's identity and foot stance while swapping the shoe. Identity and pose drift are the honest pressure points.

Face preservation

target ≥ 0.99 landmark cosine

0.998 pass

n=49 · 95% CI 0.99–1.00 · mediapipe 0.10.35 · pipeline v3.1 · set shoes-v3 · measured 2026-06-16

Pose / stance preservation

target ≤ 0.15 stance drift (torso-norm)

0.022 pass

n=49 · 95% CI 0.01–0.03 · mediapipe 0.10.35 · pipeline v3.1 · set shoes-v3 · measured 2026-06-16

Render sharpness (BRISQUE)

target threshold pending calibration

67.7

n=49 · 95% CI 66.47–68.97 · piq.brisque 0.8.0 · pipeline v3.1 · set shoes-v3 · measured 2026-06-16

Auto-QC first-pass rate

target ≥ 85% first pass

93.8% pass

n=49 · 95% CI 83.47–97.90 · pixfino-qc-gate n/a · pipeline v3.1 · set shoes-v3 · measured 2026-06-16