ZeDaq.ai

Developer Docs

Three REST endpoints, authenticated with the API key from your dashboard. The explorers below make real requests — paste in a real key from API Keys in your dashboard and try them live.

Status

All three endpoints are live and fully functional today, including POST /api/v1/tryon, which renders real Virtual Try-On images.

On Shopify? Skip the API entirely

If your store is on Shopify, none of the REST integration below is necessary — the ZeDaq.ai app installs the whole flow for you, no code required.

  1. 1. Install the app on your store (below).
  2. 2. In Shopify Admin, go to Online Store → Themes → Customize → App embeds and turn on “ZeDaq Virtual Try-On.”
  3. 3. Visit any product page — the “Try it on” button is already live, translated automatically into your storefront’s language.
  4. 4. In the embedded ZeDaq.ai admin (Shopify Admin → Apps → ZeDaq.ai), pick a plan — trial credits are included from install.

Quickstart (REST API — for everything else)

Not on Shopify, or building a custom integration? The fastest path from signup to a working integration. Each step links to the fuller reference further down this page.

  1. 1. Create your account and grab an API key

    Sign up (free trial, no card required), then copy your key from API Keys in your dashboard — you only see the full value once, at creation. Every request needs it in an x-api-key header.

  2. 2. Add size recommendations to your product pages

    Call GET /v1/size-chartwith a shopper’s measurements — free, no credits consumed. This alone is enough to show “Recommended size: M” without touching Virtual Try-On at all. See the full reference below.

  3. 3. Add Virtual Try-On

    Call POST /v1/tryonwith a photo of the shopper and a photo of your product — you get back a rendered image of the shopper wearing it. This is the highest-value endpoint and the one that costs the most credits (3/call), so it’s usually the last piece wired up, once sizing is already working.

  4. 4. (Optional) Add real body scanning for accurate sizing

    GET /v1/size-chart only works if you already know the shopper’s measurements. POST /v1/scancomputes them for you from two photos (front and side) with an A4 sheet of paper held up as a size reference — your client handles the photo capture and pose/landmark extraction (we recommend Google’s MediaPipe, which runs free in-browser), and this endpoint does the calibration math. This is the most involved endpoint to integrate — build it last, and treat it as an optional, opt-in step for shoppers, not something you gate try-on behind.

  5. 5. Before you go live

    • Create a separate labeled key per storefront if you run more than one (from API Keys) — usage shows up broken down per key, and you can revoke one without affecting the others.
    • Check your plan’s monthly call cap on the Overview page and upgrade before you expect to exceed it.
    • Resize images to roughly 1000px on the long edge before sending — see the note under POST /v1/tryon below.
    • Keep your key server-side. Call these endpoints from your own backend, never directly from browser JS — see the demo store’s source for a reference implementation of this pattern.

Authentication

Every request needs an x-api-key header set to a key from your dashboard. Keys are SHA-256 hashed at rest; we never store or display the raw value again after creation.

GET /v1/size-chart

Free — doesn’t consume credits. Pass chest_cm (and optionally waist_cm, hips_cm) to get back an alpha size and regional labels. Omit chest_cm to get the raw alpha-band table.

GET/api/v1/size-chart

POST /v1/scan

1 credit per call. Computes chest/waist/hip circumference from a front-view width and side-view depth at each level, calibrated against a visible A4 sheet in each photo (Ramanujan’s ellipse-perimeter approximation). Your client does the photo capture and pose/landmark extraction (e.g. MediaPipe) and sends the resulting pixel measurements here — this endpoint does the calibration math and sizing.

POST/api/v1/scan

POST /v1/tryon

3 credits per call. Renders your garment on the shopper’s photo using our Virtual Try-On model. Both images are base64-encoded (no data: URI prefix) — swap the placeholders below for real image data to get a real render back.

Keep each image under ~3MB (resize to roughly 1000px on the long edge before encoding) — full-resolution phone photos can push the combined request over our 4.5MB body limit and return a 413.

POST/api/v1/tryon