Anthropometric API
Size Your Customer. From height, weight,
or any dimension. No Photos. No PII. Stateless.
Supply height, weight, and any other body measurement. Get back a complete anthropometric customer's profile. Up to 130 body dimensions in one API call — ISO 7250-1 codes where available, calibrated confidence scores, and 95% prediction intervals. Not AI — peer-reviewed statistical models on real anthropometric data. GDPR and HIPAA-ready.
- 130
- Dimensions
- Full body profile per call · ISO codes where available
- 5
- Body Bundles
- Full Body
Torso
Head & Face
Hand & Arm
Legs & Feet - 3
- Build Types
- Civilian
Athletic
Overweight - 7
- Regional Profiles
- Global
Europe
Asia Pacific
Latin America
India
Africa
Middle East - 0
- PII Stored
- Stateless by architecture
- 130
- Dimensions
- Full body profile per call · ISO codes where available
- 5
- Body Bundles
- Full Body
Torso
Head & Face
Hand & Arm
Legs & Feet - 3
- Build Types
- Civilian
Athletic
Overweight - 7
- Regional Profiles
- Global
Europe
Asia Pacific
Latin America
India
Africa
Middle East - 0
- PII Stored
- Stateless by architecture
Live API Demo
Put us
to the test.
Enter your gender and height. The API instantly returns three body dimensions you can verify on the spot with a ruler right at your desk — no setup, no account needed.
In the full production API, the same single POST call returns up to 130 dimensions. Data Dictionary →
What you can build
Our data.
Your product.
One POST call returns a complete body profile — up to 130 dimensions from height and weight alone. Below are examples of the products developers ship on top of that data layer: sizing widgets, rental pre-sizers, avatar rigs, fit engines. The body dimensions come from DimensionsPot. What you build with them is entirely up to you.
No Black Box
Not AI.
Peer-reviewed science.
DimensionsPot doesn't guess. It applies peer-reviewed statistical models to ISO-standardised anthropometric survey data. Every output is traceable — to the dataset, the model, and the equation. No neural network. No training data. No probabilistic hallucination.
Every dimension comes with a confidence score, a 95% prediction interval, and biological limit flags — plus an ISO 7250-1 code where the standard defines one. If you need to know why the model returned a specific value, you can audit it. That's the difference between a statistical model and a black box.
The Standard
Every dimension maps to the international standard for basic human body measurements — the same framework used in ergonomic product certification and occupational safety.
Peer-Reviewed Datasets
Primary training data — ANSUR II, NHANES, CDC — publicly archived and citable. Regional calibration derived from peer-reviewed international anthropometric surveys.
Published Methods
Adult predictions use Ridge Regression on ANSUR II. Pediatric uses LMS Box-Cox calibrated to CDC and WHO growth standards. Both are well-understood, published statistical methods.
Explainable Output
Every dimension returns a confidence score and a 95% prediction interval. ISO 7250-1 codes are included where the standard defines them. The model never returns a number without telling you exactly how certain it is.
How it works
From a few numbers
to a full body profile.
Send a POST request
Send a POST request
Supply gender and at least one measurement. Height and weight together give the best accuracy — but a single anchor is sufficient. Metric and imperial both accepted.
body_height: 1780mm
body_mass: 82.0kg Dual-Core Inference runs
Dual-Core Inference runs
The Adult Ridge Regression engine (ANSUR II) or Pediatric LMS Box-Cox model (CDC/WHO) generates predictions through a 9-step pipeline. Missing anchors are reconstructed via dynamic imputation.
9-step pipeline
Deterministic · Stateless Receive 130 dimensions
Receive 130 dimensions
Each dimension includes a value, type (BONE/FLESH), Confidence Score, biological limit status, and optional 95% prediction interval with ISO 7250-1 code.
confidence_score: 85
range_95: [954, 1070] Capabilities
Built for production,
designed for privacy.
Stateless by Architecture
No PII ever stored, logged, or retained between calls. Data is processed in-memory and discarded. GDPR and HIPAA-ready not as a policy, but as a structural fact.
Privacy Architecture130 ISO 7250-1 Dimensions
Skeletal landmarks, soft-tissue measurements, body composition — defined using ISO 7250-1 anatomical methodology, with standard codes where the ISO specification assigns them. Directly usable for product certification and ergonomic design.
Data DictionaryBody Bundles
Request only what you need — FULL_BODY (130 dimensions), or named subsets: HEAD_FACE, HAND_ARM, TORSO, LEGS_FEET. Reduces payload size and simplifies downstream processing.
Bundle ReferenceCalibrated Confidence Score
Every output dimension carries a Confidence Score [0–100] and an optional 95% prediction interval. The system never over-promises — actual coverage ≥ stated score.
Confidence Score7 Regional Profiles
Separate calibration for GLOBAL, EUROPE, ASIA_PACIFIC, LATAM, INDIA, AFRICA, and MIDDLE_EAST. Input origin and output target are fully independent fields.
Regional CalibrationBody Build Types
Three morphological presets — CIVILIAN (general population), ATHLETIC (lean, reduced soft-tissue shift), and OVERWEIGHT (BMI-stratified NHANES morphing) — applied before inference to match your customer profile.
Build TypesAnchor Tiers & Confidence
The more you supply,
the more precise
the result.
Every dimension carries a Confidence Score [0–100]. BONE skeletal landmarks are more predictable than FLESH soft-tissue measurements. The system is calibrated to never over-promise — stated coverage ≥ actual coverage.
Each output also optionally returns a 95% prediction interval (range_95) — the statistical range within which the true measurement falls for 95% of individuals with the same inputs. Useful for tolerance stacking, sizing logic, and quality control.
PRIMARY_RICH Height + weight + ≥1 circumference
PRIMARY_BOTH Height + weight
PRIMARY_ONE Height OR weight only
SECONDARY Foot length, knee height…
TERTIARY Any other single measurement
| Tier | Input supplied | BONE | FLESH |
|---|---|---|---|
PRIMARY_RICH | Height + weight + ≥1 circumference | ~87 | ~80 |
PRIMARY_BOTH | Height + weight | ~85 | ~78 |
PRIMARY_ONE | Height OR weight only | ~79 | ~62 |
SECONDARY | Foot length, knee height… | ~74 | ~67 |
TERTIARY | Any other single measurement | ~69 | ~62 |
Pediatric Module — included
Children aren't
small adults.
A dedicated Pediatric LMS Box-Cox model calibrated against CDC and WHO growth standards. Unlike adult regression, pediatric inference accounts for non-linear growth curves across each developmental stage — with separate confidence scoring at every tier.
Age or age category alone is sufficient — no body measurements required
Separate confidence scoring calibrated to pediatric growth variability
Stateless architecture — no PII stored, COPPA and GDPR-ready by design
INFANT WHO Multicentre Growth Reference Study. Non-linear growth at monthly granularity.
TODDLER Rapid proportional shifts in trunk and limb ratios accounted for at each age step.
CHILD CDC growth charts with LMS transformation. Steady mid-childhood growth curve.
PRE_TEEN Pre-pubertal acceleration phase modelled separately for each gender.
TEEN WHO adolescent standards with NHANES calibration. Puberty-stage growth modelled.
Use Cases
Built for every vertical
that needs body data.
Fashion & Apparel E-commerce
Gaming, VFX & Metaverse
Online Eyewear & VR/AR Headsets
Sport Equipment & Outdoor Gear Rental
Wearables & Smart Accessories
Childrenswear & Children's Products
Workwear at Scale
Simple to Integrate · Private by Architecture
DimensionsPot vs
Photo-Based Sizing
Two numeric inputs. Minutes to integrate. No photos, no biometric data, no GDPR special categories. The simplicity and privacy advantage is structural — not a policy.
Height + weight (optional: any circumference). Two numeric fields.
2 full-body photos, strict lighting & pose required
Nothing — stateless by architecture. No logs, no retention, no user profiles.
Photos retained for model processing and retraining
No biometric data. No GDPR Article 9 obligations. No EU AI Act biometric classification.
Photos = GDPR Art. 9 biometric data. Strict EU AI Act obligations apply.
One POST request. 2 numeric input fields. Any language — no SDK required.
Photo upload flow, CDN, storage, moderation, pose-validation layer
Minutes — subscribe on RapidAPI, copy key, send first request.
Days to weeks — build photo infra, consent UX, moderation pipeline
None required. Standard HTTPS POST — no storage, no media pipeline.
Storage layer, CDN, image processing, model hosting required
Photographs are biometric data under GDPR Article 9 — a special category requiring explicit legal basis and strict safeguards. Any link between a body profile and a specific individual exists exclusively in your own infrastructure.
Pricing
Start free.
Scale when you're ready.
All features available on every tier. No feature gating — plans differ only in monthly request volume. A practical alternative to enterprise photo-based sizing platforms.
Think in value, not volume
The right unit isn't cost per API call — it's value per successful inference. A call that prevents a returned garment, avoids a mis-sized PPE order, or converts an undecided buyer recovers its cost before the month ends.
The API is deterministic — same inputs always return the same output. Cache aggressively client-side: one call per user profile, not one call per session.
Costs you avoid
Need unlimited calls or on-premise deployment? Contact us for Enterprise →
From the Blog
Latest articles
Build vs. Buy: Adding Size Recommendations to Your App Without Training a Model
Build vs. buy for size recommendation engines — data requirements, ML infrastructure, accuracy tradeoffs, and time-to-production framework for developer teams.
How to Store Body Measurements: Database Schema Design
Database schema design for body measurement applications — what to persist, what to compute on demand, privacy-by-design patterns, and temporal versioning.
How to Add Body Measurements to Your App Without Photos or Privacy Risk
Photo-based sizing creates GDPR exposure and user friction. How statistical body measurement prediction works as a privacy-first alternative, with Python code.
All articles
Explore the full Knowledge Base
Build vs. Buy: Adding Size Recommendations to Your App Without Training a Model
Build vs. buy for size recommendation engines — data requirements, ML infrastructure, accuracy tradeoffs, and time-to-production framework for developer teams.
How to Store Body Measurements: Database Schema Design
Database schema design for body measurement applications — what to persist, what to compute on demand, privacy-by-design patterns, and temporal versioning.
How to Add Body Measurements to Your App Without Photos or Privacy Risk
Photo-based sizing creates GDPR exposure and user friction. How statistical body measurement prediction works as a privacy-first alternative, with Python code.
Start building
in under 5 minutes.
Subscribe on RapidAPI, use the pre-filled Playground example, and receive your first 130-dimension body profile. No SDK required.
Disclaimer: All outputs of the DimensionsPot API are statistically derived anthropometric predictions intended to support — not replace — professional judgment. They do not constitute medical, clinical, ergonomic, or professional advice. The Confidence Score is a proprietary heuristic index — not a statistical confidence interval. To the fullest extent permitted by applicable law, DimensionsPot disclaims all liability for any damages arising from reliance on API outputs.