Noisy Student Training using Body Language Dataset Improves Facial Expression Recognition

Abstract

Facial expression recognition from videos in the wild is a challenging task due to the lack of abundant labelled training data. Large DNN (deep neural network) architectures and ensemble methods have resulted in better performance, but soon reach saturation at some point due to data inadequacy. In this paper, we use a self-training method that utilizes a combination of a labelled dataset and an unlabelled dataset (Body Language Dataset - BoLD). Experimental analysis shows that training a noisy student network iteratively helps in achieving significantly better results. Additionally, our model isolates different regions of the face and processes them independently using a multi-level attention mechanism which further boosts the performance. Our results show that the proposed method achieves state-of-the-art performance on benchmark datasets CK+ and AFEW 8.0 when compared to other single models.

arXiv commentAccepted in ECCV workshop BEEU (Bodily Expressed Emotion Understanding) 2020

References in corpus (9)

• Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks
• A Structured Self-attentive Sentence Embedding
• Deep Learning using Linear Support Vector Machines
• Semi-Supervised Learning with Ladder Networks
• Learning with Pseudo-Ensembles
• Aff-Wild2: Extending the Aff-Wild Database for Affect Recognition
• ARBEE: Towards Automated Recognition of Bodily Expression of Emotion In the Wild
• Convolutional neural networks pretrained on large face recognition datasets for emotion classification from video
• Emotion Recognition with Spatial Attention and Temporal Softmax Pooling

Cited by in corpus (3)

• Facial expression and attributes recognition based on multi-task learning of lightweight neural networks
• Mutual Information Regularized Identity-aware Facial ExpressionRecognition in Compressed Video
• OmniLayout: Room Layout Reconstruction from Indoor Spherical Panoramas