Robust Head-Pose Estimation Based on Partially-Latent Mixture of Linear Regressions

TL;DR

This paper introduces a robust head-pose estimation method using a mixture of linear regressions with partially-latent outputs, effectively handling occlusions, illumination changes, and alignment errors across diverse datasets.

Contribution

It presents a novel regression approach combining manifold learning and mixtures to improve head-pose estimation robustness under challenging conditions.

Findings

Outperforms several state-of-the-art methods on public datasets.

Demonstrates robustness to occlusions and alignment errors.

Validates effectiveness across multiple datasets.

Abstract

Head-pose estimation has many applications, such as social event analysis, human-robot and human-computer interaction, driving assistance, and so forth. Head-pose estimation is challenging because it must cope with changing illumination conditions, variabilities in face orientation and in appearance, partial occlusions of facial landmarks, as well as bounding-box-to-face alignment errors. We propose tu use a mixture of linear regressions with partially-latent output. This regression method learns to map high-dimensional feature vectors (extracted from bounding boxes of faces) onto the joint space of head-pose angles and bounding-box shifts, such that they are robustly predicted in the presence of unobservable phenomena. We describe in detail the mapping method that combines the merits of unsupervised manifold learning techniques and of mixtures of regressions. We validate our method with three publicly available datasets and we thoroughly benchmark four variants of the proposed algorithm with several state-of-the-art head-pose estimation methods.