Body Measurement API Providers Compared

Body measurement API providers either measure a person from photos or predict their measurements from a few answers. 3DLOOK measures from two phone photos and returns over 80 measurements, Bold Metrics predicts a body profile from shopper inputs with no photos, and ZeDaq.ai measures from one photo plus height and maps the result to a store's size chart.

Which one fits depends on what you need the numbers for: a size label, a made-to-measure pattern, or health and fitness tracking. If your project also needs try-on, start with our developer guide to try-on APIs.

Vendor facts here come from each provider's own site, checked November 2026. Read current API docs before building.

How do the providers compare at a glance?

This table uses only what each provider's live site says.

	3DLOOK	Bold Metrics	ZeDaq.ai
Method	Photo scan	Prediction from answers	Photo scan
Input	Two photos, front and side, in fitted clothing	4 to 6 inputs such as height, weight, age	One photo plus height
Output	80+ measurements, 3D body model	Digital twin of 50+ measurements, size and fit rating	Measurements matched to the store's size chart
Main uses named	Made-to-measure, uniforms, health and fitness	Ready-to-wear sizing, custom clothiers, AI shopping agents	Fashion store sizing and try-on
Developer access	API/SDK; Android and iOS camera SDKs	Virtual Sizer API	REST API and script-tag widget
Compliance named	HIPAA under a BAA, GDPR	Not stated on homepage	Not covered here
Public pricing	Not shown; 7-day Mobile Tailor trial	Not shown; demo	Free to $299/mo

Sources: 3DLOOK homepage, 3DLOOK Mobile Tailor and Bold Metrics homepage, checked November 2026.

What does a body measurement API actually return?

It depends on the provider, but outputs fall into three layers.

1. Raw measurements: numbers such as chest, waist, hips, inseam and shoulder width.
2. A body model: a 3D shape or digital twin built from those numbers.
3. A decision: a size label, a fit rating or a flag for your team.

Some providers stop at layer one or two and leave the size decision to you. Others go straight to layer three. Know which you need before you compare.

How does 3DLOOK measure bodies?

3DLOOK scans the person from two photos taken through its own link. Its Mobile Tailor page describes voice instructions and a gyroscope that guide the front and side photos.

What it returns

The Mobile Tailor page mentions over 80 measurements in one section and over 85 in another, plus a 3D body model. Results arrive in under 60 seconds in one section and under 3 minutes for the full scan in the FAQ.

Accuracy claims

3DLOOK's homepage states 96–97% accuracy for body measurements on its FitXpress product, according to 3DLOOK. Treat any vendor accuracy figure as a claim to test, and ask how it was measured.

Health and compliance

The homepage leads with FitXpress for health and fitness, with use cases such as telehealth and insurance. It says HIPAA deployments are possible under a business associate agreement and that GDPR is supported, per 3DLOOK.

How does Bold Metrics estimate bodies?

Bold Metrics doesn't take photos. Its homepage says it uses "4 – 6 simple shopper inputs", such as height, weight, age, pant waist, fit preference, bra band and cup size.

From those, it builds a digital twin of "50+" body measurements. For stores, it returns a best-fit size per style with a fit rating. For custom clothiers, it names Virtual Tailor, which provides tailor-level body measurements.

The appeal is low effort for the shopper. The question to ask is how well predicted measurements match real ones for shoppers with less common body shapes.

How does ZeDaq.ai measure bodies?

ZeDaq.ai uses a 30-second scan with one photo plus height. The height gives the scan a real-world scale, so the photo can be turned into body measurements.

Those measurements are matched to the store's own size chart. Merchants enter one chart per garment type, such as tops, outerwear, dresses and bottoms, and the scan recommends the store's labels. It sizes up to the largest relevant measurement, so a shopper with broad hips isn't put in a size that's tight there.

If no chart is set, a generic UK, US, EU or AU chart is used. See how AI size recommendation works for the shopper side, and our ZeDaq.ai vs 3DLOOK scan comparison for a closer look at the two scan methods.

Which provider fits which use?

Your use	Better fit	Reason
Custom suits or bridal patterns	3DLOOK	Many measurements from a guided scan
Health or fitness app	3DLOOK	FitXpress and HIPAA under a BAA are named
Ready-to-wear store, camera-shy shoppers	Bold Metrics	No photos needed
Fashion store wanting size plus try-on	ZeDaq.ai	One photo for sizing, one for try-on
Staff uniform fitting at scale	3DLOOK or ZeDaq.ai	Remote scan links or a widget

If you only need a size label, not raw measurements, our comparison of sizing API providers covers more widget-first tools.

How should you test measurement accuracy yourself?

Vendor accuracy numbers are a start, but your own test is what counts. Here's a simple method.

1. Recruit ten people with a range of heights and builds. Include staff of different sizes.
2. Take tape measurements of chest, waist and hips, using the same person to measure everyone.
3. Run each provider's scan or form for every person, following the provider's instructions.
4. Record the difference for each measurement, in centimeters or inches.
5. Look at the spread, not just the average. A tool that's close on average but far off for some people may cause more returns.

A worked example

Say a tester has a tape-measured waist of 84 cm. Provider A returns 85 cm, and Provider B returns 90 cm. On a chart where size M covers waist 80 to 86 cm, Provider A suggests M and Provider B suggests L.

That one test shows how a 6 cm gap can change the size. Run the same check across all ten testers before deciding.

How do you get good scans from shoppers?

A photo-based API is only as good as the photos it gets. Most measurement errors start with the capture, not the model. Your on-page instructions should cover these points:

• Clothing: ask for fitted clothes. 3DLOOK's Mobile Tailor page, for example, suggests leggings and a fitted top.
• Pose: show a simple drawing of the pose, with arms slightly away from the body.
• Framing: the whole body, head to feet, should be in the photo.
• Background: a plain wall works better than a busy room.
• Height: if the API asks for height, remind shoppers to enter it without shoes.

What should happen when a scan fails?

Plan for failed scans from the start. If the photo is too dark or the person is cut off, show a short, friendly message with one fix, such as "Step back so your feet are in the photo." Then let the shopper try again or skip to the size chart.

Never leave a shopper stuck without a size answer. A fallback to your written size chart keeps the sale moving.

How do answer-based tools handle errors?

Answer-based tools fail differently. A shopper may type weight in the wrong unit or skip a question. Check that the provider validates inputs and asks again when an answer looks wrong, rather than returning a size from bad data.

What should you ask about data and privacy?

Body measurements and photos are personal data. Ask every provider these questions:

• Are photos processed on the device, on your servers or both?
• How long are photos and measurements stored?
• Can a person ask for their data to be deleted?
• Is any data used to train your models?
• Which laws and frameworks do you support, and can you sign a data agreement?

Get the answers in writing and keep them with your contract.

How is ZeDaq.ai priced?

The Free plan is $0, with 3 try-ons, 3 measurements and 5 invoices a month. Paid plans are Micro $14, Starter $49, Growth $129 and Scale $299 a month, each with a 3-day free trial and no card required.

FAQs

Can a body measurement API work without photos?

Yes. Bold Metrics describes building a body profile from a few shopper answers, such as height, weight and age, with no photos.

How many measurements does 3DLOOK return?

Its Mobile Tailor page says over 80 measurements in one section and over 85 in another, plus a 3D body model, checked November 2026.

Why does ZeDaq.ai ask for height?

Height gives the scan a real-world scale. Without it, a photo can show body proportions but not true sizes in centimeters or inches.

Are vendor accuracy figures reliable?

Treat them as claims. Run your own tape-measure test with a range of people before you choose a provider.