The FaceChannel: A Light-weight Deep Neural Network for Facial Expression Recognition
This work addresses the challenge of adapting FER models to changing conditions, which is important for applications in affect perception, but it is incremental as it focuses on reducing model complexity rather than a fundamental breakthrough.
The paper tackles the problem of training deep neural networks for facial expression recognition (FER) by proposing the FaceChannel, a light-weight network with fewer parameters, which achieves comparable or better performance on benchmark datasets compared to state-of-the-art models.
Current state-of-the-art models for automatic FER are based on very deep neural networks that are difficult to train. This makes it challenging to adapt these models to changing conditions, a requirement from FER models given the subjective nature of affect perception and understanding. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We perform a series of experiments on different benchmark datasets to demonstrate how the FaceChannel achieves a comparable, if not better, performance, as compared to the current state-of-the-art in FER.