Seminar: Graduate Seminar
Posture-Aware Age Group Estimation from Smartphone-Based 3D Scans Using PCA
Date:
June,17,2025
Start Time:
10:00 - 11:00
Location:
Taub 401
Add to:
Lecturer:
Asaf Levy
Research Areas:
| This study presents a method for estimating the age group of individuals based on 3D upper- body scans, focusing on spinal posture as a biologically informative trait. Using smartphone-generated 3D scans, we sample consistently ordered anatomical points from the spine region via a structured grid, following normalization with anatomical landmarks. We then apply a regularized Principal Component Analysis (RPCA) framework that enforces spatial smoothness through a Laplacian prior. The resulting low-dimensional embedding reveals a clear separation between younger and older subjects, demonstrating the framework’s ability to capture age-related morphological variation in an unsupervised manner. These findings support the potential of accessible smartphone-based scanning technology for extracting posture-related biomarkers, with broader implications for non-invasive biometric analysis in future work.
M.Sc. student under the supervision of Prof. Ron Kimmel .
|