Pose-independent 3D Anthropometry from Sparse Data

2025-01-10

AI Summary

To address the challenge of acquiring accurate 3D anthropometric measurements from individuals with limited mobility—who cannot reliably assume the standard “A-pose”—this paper proposes the first pose-invariant, sparse-input 3D human measurement framework. Given only a single frame of 2D/3D skeletal keypoints under arbitrary pose, our method performs geometric normalization and learns pose-size-decoupled features to directly regress 12 core body measurements under the canonical A-pose in an end-to-end manner. Key contributions include: (i) the first anthropometric framework eliminating both A-pose constraints and reliance on dense geometric inputs (e.g., meshes or point clouds); (ii) a generalizable sparse representation applicable to arbitrary poses; and (iii) the first open, ready-to-use 3D anthropometric benchmark framework. Evaluated on public datasets, our method achieves a mean error of <1.2 cm—comparable to state-of-the-art dense-scanning approaches—while significantly improving accessibility and inclusivity for injured, elderly, or disabled populations.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Humans and AI: AI for Accessibility
• Intelligent Robots: State Estimation

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

3D digital anthropometry is the study of estimating human body measurements from 3D scans. Precise body measurements are important health indicators in the medical industry, and guiding factors in the fashion, ergonomic and entertainment industries. The measuring protocol consists of scanning the whole subject in the static A-pose, which is maintained without breathing or movement during the scanning process. However, the A-pose is not easy to maintain during the whole scanning process, which can last even up to a couple of minutes. This constraint affects the final quality of the scan, which in turn affects the accuracy of the estimated body measurements obtained from methods that rely on dense geometric data. Additionally, this constraint makes it impossible to develop a digital anthropometry method for subjects unable to assume the A-pose, such as those with injuries or disabilities. We propose a method that can obtain body measurements from sparse landmarks acquired in any pose. We make use of the sparse landmarks of the posed subject to create pose-independent features, and train a network to predict the body measurements as taken from the standard A-pose. We show that our method achieves comparable results to competing methods that use dense geometry in the standard A-pose, but has the capability of estimating the body measurements from any pose using sparse landmarks only. Finally, we address the lack of open-source 3D anthropometry methods by making our method available to the research community at https://github.com/DavidBoja/pose-independent-anthropometry.

Problem

• 3D anthropometry
• posture-unrestricted measurement
• limited data

Innovation

• 3D body measurement
• pose-invariant features
• open-source methodology