Unsupervised Features for Facial Expression Intensity Estimation over Time

TL;DR

This paper introduces an unsupervised, person-invariant feature for estimating facial expression intensity over time, outperforming existing methods and applicable to various tasks like sequence alignment and action unit analysis.

Contribution

The authors propose a novel dynamic feature for facial expression analysis that is invariant to individual differences and expression types, improving robustness and versatility.

Findings

Outperforms state-of-the-art in expression intensity estimation

Effective in temporal alignment of facial sequences

Robust against noise and outliers

Abstract

The diversity of facial shapes and motions among persons is one of the greatest challenges for automatic analysis of facial expressions. In this paper, we propose a feature describing expression intensity over time, while being invariant to person and the type of performed expression. Our feature is a weighted combination of the dynamics of multiple points adapted to the overall expression trajectory. We evaluate our method on several tasks all related to temporal analysis of facial expression. The proposed feature is compared to a state-of-the-art method for expression intensity estimation, which it outperforms. We use our proposed feature to temporally align multiple sequences of recorded 3D facial expressions. Furthermore, we show how our feature can be used to reveal person-specific differences in performances of facial expressions. Additionally, we apply our feature to identify the local changes in face video sequences based on action unit labels. For all the experiments our feature proves to be robust against noise and outliers, making it applicable to a variety of applications for analysis of facial movements.