Towards Unbiased Label Distribution Learning for Facial Pose Estimation Using Anisotropic Spherical Gaussian
This work addresses bias and inflexibility in facial pose estimation, an incremental improvement for computer vision applications like human-computer interaction.
The paper tackled biased label distribution learning in facial pose estimation by proposing an Anisotropic Spherical Gaussian approach that generates unbiased expectations and learns flexible parameters per sample, achieving new state-of-the-art results on AFLW2000 and BIWI datasets.
Facial pose estimation refers to the task of predicting face orientation from a single RGB image. It is an important research topic with a wide range of applications in computer vision. Label distribution learning (LDL) based methods have been recently proposed for facial pose estimation, which achieve promising results. However, there are two major issues in existing LDL methods. First, the expectations of label distributions are biased, leading to a biased pose estimation. Second, fixed distribution parameters are applied for all learning samples, severely limiting the model capability. In this paper, we propose an Anisotropic Spherical Gaussian (ASG)-based LDL approach for facial pose estimation. In particular, our approach adopts the spherical Gaussian distribution on a unit sphere which constantly generates unbiased expectation. Meanwhile, we introduce a new loss function that allows the network to learn the distribution parameter for each learning sample flexibly. Extensive experimental results show that our method sets new state-of-the-art records on AFLW2000 and BIWI datasets.