Continual Facial Expression Recognition: A Benchmark

TL;DR

This paper introduces the ConFER benchmark to evaluate continual learning methods for facial expression recognition, demonstrating their effectiveness in adapting to real-world, incremental data while avoiding catastrophic forgetting.

Contribution

It proposes a new benchmark for continual learning in FER, compares various CL techniques on multiple datasets, and discusses their potential and challenges in affective computing.

Findings

CL techniques achieve SOTA performance on FER datasets

Continual learning helps models adapt to real-world, incremental data

Strategies for effective implementation of CL in FER are discussed

Abstract

Understanding human affective behaviour, especially in the dynamics of real-world settings, requires Facial Expression Recognition (FER) models to continuously adapt to individual differences in user expression, contextual attributions, and the environment. Current (deep) Machine Learning (ML)-based FER approaches pre-trained in isolation on benchmark datasets fail to capture the nuances of real-world interactions where data is available only incrementally, acquired by the agent or robot during interactions. New learning comes at the cost of previous knowledge, resulting in catastrophic forgetting. Lifelong or Continual Learning (CL), on the other hand, enables adaptability in agents by being sensitive to changing data distributions, integrating new information without interfering with previously learnt knowledge. Positing CL as an effective learning paradigm for FER, this work presents the Continual Facial Expression Recognition (ConFER) benchmark that evaluates popular CL techniques on FER tasks. It presents a comparative analysis of several CL-based approaches on popular FER datasets such as CK+, RAF-DB, and AffectNet and present strategies for a successful implementation of ConFER for Affective Computing (AC) research. CL techniques, under different learning settings, are shown to achieve state-of-the-art (SOTA) performance across several datasets, thus motivating a discussion on the benefits of applying CL principles towards human behaviour understanding, particularly from facial expressions, as well the challenges entailed.