Feature Super-Resolution Based Facial Expression Recognition for Multi-scale Low-Resolution Faces

TL;DR

This paper introduces a novel feature super-resolution method using GANs to improve facial expression recognition on low-resolution images, outperforming existing image super-resolution and recognition techniques.

Contribution

A new feature-level super-resolution approach using GANs combined with a classification-aware loss for robust FER on very low-resolution images.

Findings

Achieves better FER accuracy on low-res images than existing methods.

Effective across various down-sampling factors with a single model.

Demonstrates improved discriminative features for expression recognition.

Abstract

Facial Expressions Recognition(FER) on low-resolution images is necessary for applications like group expression recognition in crowd scenarios(station, classroom etc.). Classifying a small size facial image into the right expression category is still a challenging task. The main cause of this problem is the loss of discriminative feature due to reduced resolution. Super-resolution method is often used to enhance low-resolution images, but the performance on FER task is limited when on images of very low resolution. In this work, inspired by feature super-resolution methods for object detection, we proposed a novel generative adversary network-based feature level super-resolution method for robust facial expression recognition(FSR-FER). In particular, a pre-trained FER model was employed as feature extractor, and a generator network G and a discriminator network D are trained with features extracted from images of low resolution and original high resolution. Generator network G tries to transform features of low-resolution images to more discriminative ones by making them closer to the ones of corresponding high-resolution images. For better classification performance, we also proposed an effective classification-aware loss re-weighting strategy based on the classification probability calculated by a fixed FER model to make our model focus more on samples that are easily misclassified. Experiment results on Real-World Affective Faces (RAF) Database demonstrate that our method achieves satisfying results on various down-sample factors with a single model and has better performance on low-resolution images compared with methods using image super-resolution and expression recognition separately.