Full-range Head Pose Geometric Data Augmentations

TL;DR

This paper introduces novel geometric data augmentation methods for head pose estimation that improve accuracy across the full range of head angles by addressing coordinate system and Euler angle ambiguities.

Contribution

It presents new formulas and derivations for accurate 2D geometric augmentations and coordinate system inference, enhancing full-range head pose dataset generation.

Findings

Improved head pose estimation accuracy across full angle range

Enhanced dataset generation with accurate rotation matrix augmentations

Significant performance boost in existing models using proposed methods

Abstract

Many head pose estimation (HPE) methods promise the ability to create full-range datasets, theoretically allowing the estimation of the rotation and positioning of the head from various angles. However, these methods are only accurate within a range of head angles; exceeding this specific range led to significant inaccuracies. This is dominantly explained by unclear specificity of the coordinate systems and Euler Angles used in the foundational rotation matrix calculations. Here, we addressed these limitations by presenting (1) methods that accurately infer the correct coordinate system and Euler angles in the correct axis-sequence, (2) novel formulae for 2D geometric augmentations of the rotation matrices under the (SPECIFIC) coordinate system, (3) derivations for the correct drawing routines for rotation matrices and poses, and (4) mathematical experimentation and verification that allow proper pitch-yaw coverage for full-range head pose dataset generation. Performing our augmentation techniques to existing head pose estimation methods demonstrated a significant improvement to the model performance. Code will be released upon paper acceptance.