Applied Vision

Synthetic Skeletal Pose Pre-training to Mitigate Data Scarcity in In-Cabin 2D-to-3D Pose Lifting

Authors: Thummanoon Kunanuntakij (TU Wien) , Dominik Schörkhuber (TU Wien) , Margrit Gelautz (TU Wien)

Abstract

Driver-related factors contribute to nearly 90% of traffic accidents. Estimating 3D driver poses can help track risky behaviors. However, the scarcity of annotated 3D pose data, together with the complexity and high cost of 3D annotation, limits the training of domain-specific estimators. We address this challenge by pre-training 2D-to-3D pose lifting models using synthetic 3D poses from a simulated dataset. In experiments on the Drive&Act dataset, we compare training from scratch with synthetic pre-training while gradually increasing the amount of real-world data. For example, when only 5% of training data is available, MPJPE is reduced from 90.0 mm to 70.9 mm for the GraFormer model. Our results demonstrate that synthetic pre-training consistently reduces estimation errors, particularly when real-world data are limited. Furthermore, synthetic pre-training improves the best fine-tuned results across different models from 48.1 mm to 46.0 mm in our tests.

Keywords:

How to Cite: Kunanuntakij, T. , Schörkhuber, D. & Gelautz, M. (2026) “Synthetic Skeletal Pose Pre-training to Mitigate Data Scarcity in In-Cabin 2D-to-3D Pose Lifting”, Proceedings of the Austrian Symposium on AI, Robotics, and Vision. 3(1).