To main content

Capturing Realistic Human Body 3D Data - An Open Dataset and Pipeline for Absolute Anthropometric Measurement

Abstract

In order to make realistic and individualized digital human body models, absolute real-world measurements are needed. We share an open dataset consisting of 360° 3D scans, including raw point cloud data, 2D silhouette images and anthropometrics from twelve individuals. The dataset spans a wide range of human shapes and body mass indexes. We present a portable 3D imaging rig that captures 360° point clouds of humans at up to 30 frames per second. The system is designed to balance data quality, acquisition speed, cost and ease of setup and use. We introduce an automated online extrinsic calibration using jointly observed landmarks to simultaneously calibrate all 3D cameras and tools to perform anthropometric measurements from the point cloud data.

Category

Conference lecture

Language

English

Affiliation

  • SINTEF Digital / Health Research
  • SINTEF Digital / Smart Sensors and Microsystems
  • University of Oslo

Presented at

3DbodyTech Conference and Expo

Place

Lugano

Date

20.10.2026 - 21.10.2026

Organizer

Hometrica Consulting

Date

21.10.2026

Year

2026

View this publication at Norwegian Research Information Repository