Age-Inclusive 3D Human Mesh Recovery for Action-Preserving Data Anonymization

2025-12-04

AI Summary

Existing 3D human reconstruction methods exhibit severe generalization failure on children and infants. To address this, we propose AionHMR—a novel framework introducing SMPL-A, the first unified parametric body model jointly representing adults, children, and infants. We further construct 3D-BabyRobot, the first large-scale 3D reconstruction dataset tailored to child-robot interaction, augmented with high-fidelity pseudo-ground-truth annotations. Our method adopts an optimization-Transformer hybrid paradigm for cross-age reconstruction, balancing accuracy and real-time inference. Quantitatively, AionHMR maintains high adult reconstruction accuracy (MPJPE ↓2.1 mm) while significantly improving pose estimation for children and infants (MPJPE reduced by 37.6% and 45.3%, respectively). Additionally, it enables motion-preserving, mesh-level data anonymization for privacy protection. This work advances age-inclusive, privacy-aware 3D human modeling—bridging critical gaps in pediatric and developmental robotics applications.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: 3D Computer Vision

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Agentic search

Abstract

While three-dimensional (3D) shape and pose estimation is a highly researched area that has yielded significant advances, the resulting methods, despite performing well for the adult population, generally fail to generalize effectively to children and infants. This paper addresses this challenge by introducing AionHMR, a comprehensive framework designed to bridge this domain gap. We propose an optimization-based method that extends a top-performing model by incorporating the SMPL-A body model, enabling the concurrent and accurate modeling of adults, children, and infants. Leveraging this approach, we generated pseudo-ground-truth annotations for publicly available child and infant image databases. Using these new training data, we then developed and trained a specialized transformer-based deep learning model capable of real-time 3D age-inclusive human reconstruction. Extensive experiments demonstrate that our methods significantly improve shape and pose estimation for children and infants without compromising accuracy on adults. Importantly, our reconstructed meshes serve as privacy-preserving substitutes for raw images, retaining essential action, pose, and geometry information while enabling anonymized datasets release. As a demonstration, we introduce the 3D-BabyRobot dataset, a collection of action-preserving 3D reconstructions of children interacting with robots. This work bridges a crucial domain gap and establishes a foundation for inclusive, privacy-aware, and age-diverse 3D human modeling.

Problem

• Improves 3D shape and pose estimation for children and infants.
• Enables real-time, age-inclusive human reconstruction using a transformer model.
• Provides privacy-preserving 3D meshes as anonymized substitutes for raw images.

Innovation

• Extends adult model with SMPL-A for all ages
• Generates pseudo-ground-truth for child and infant images
• Trains transformer model for real-time age-inclusive reconstruction