Guided Spatial Transformers for Facial Expression Recognition

Facial Expression Recognition Robust to Occlusion and to Intra-Similarity Problem Using Relevant Subsampling

This paper proposes facial expression recognition (FER) with the wild data set. In particular, this paper chiefly deals with two issues, occlusion and intra-similarity problems. The attention mechanism enables one to use the most relevant areas of facial images for specific expressions, and the triplet loss function solves the intra-similarity problem that sometimes fails to aggregate the same expression from different faces and vice versa. The proposed approach for the FER is robust to occlusion, and it uses a spatial transformer network (STN) with an attention mechanism to utilize specific facial region that dominantly contributes (or that is the most relevant) to particular facial expressions, e.g., anger, contempt, disgust, fear, joy, sadness, and surprise. In addition, the STN model is connected to the triplet loss function to improve the recognition rate which outperforms the existing approaches that employ cross-entropy or other approaches using only deep neural networks or classical methods. The triplet loss module alleviates limitations of the intra-similarity problem, leading to further improvement of the classification. Experimental results are provided to substantiate the proposed approach for FER, and the result outperforms the recognition rate in more practical cases, e.g., occlusion. The quantitative result provides FER results with more than 2.09% higher accuracy compared to the existing FER results in CK+ data sets and 0.48% higher than the accuracy of the results with the modified ResNet model in the FER2013 data set.

Data Augmentation for Audio-Visual Emotion Recognition with an Efficient Multimodal Conditional GAN

Audio-visual emotion recognition is the research of identifying human emotional states by combining the audio modality and the visual modality simultaneously, which plays an important role in intelligent human-machine interactions. With the help of deep learning, previous works have made great progress for audio-visual emotion recognition. However, these deep learning methods often require a large amount of data for training. In reality, data acquisition is difficult and expensive, especially for the multimodal data with different modalities. As a result, the training data may be in the low-data regime, which cannot be effectively used for deep learning. In addition, class imbalance may occur in the emotional data, which can further degrade the performance of audio-visual emotion recognition. To address these problems, we propose an efficient data augmentation framework by designing a multimodal conditional generative adversarial network (GAN) for audio-visual emotion recognition. Specifically, we design generators and discriminators for audio and visual modalities. The category information is used as their shared input to make sure our GAN can generate fake data of different categories. In addition, the high dependence between the audio modality and the visual modality in the generated multimodal data is modeled based on Hirschfeld-Gebelein-Rényi (HGR) maximal correlation. In this way, we relate different modalities in the generated data to approximate the real data. Then, the generated data are used to augment our data manifold. We further apply our approach to deal with the problem of class imbalance. To the best of our knowledge, this is the first work to propose a data augmentation strategy with a multimodal conditional GAN for audio-visual emotion recognition. We conduct a series of experiments on three public multimodal datasets, including eNTERFACE’05, RAVDESS, and CMEW. The results indicate that our multimodal conditional GAN has high effectiveness for data augmentation of audio-visual emotion recognition.

Multimodal Emotion Recognition on RAVDESS Dataset Using Transfer Learning

Emotion Recognition is attracting the attention of the research community due to the multiple areas where it can be applied, such as in healthcare or in road safety systems. In this paper, we propose a multimodal emotion recognition system that relies on speech and facial information. For the speech-based modality, we evaluated several transfer-learning techniques, more specifically, embedding extraction and Fine-Tuning. The best accuracy results were achieved when we fine-tuned the CNN-14 of the PANNs framework, confirming that the training was more robust when it did not start from scratch and the tasks were similar. Regarding the facial emotion recognizers, we propose a framework that consists of a pre-trained Spatial Transformer Network on saliency maps and facial images followed by a bi-LSTM with an attention mechanism. The error analysis reported that the frame-based systems could present some problems when they were used directly to solve a video-based task despite the domain adaptation, which opens a new line of research to discover new ways to correct this mismatch and take advantage of the embedded knowledge of these pre-trained models. Finally, from the combination of these two modalities with a late fusion strategy, we achieved 80.08% accuracy on the RAVDESS dataset on a subject-wise 5-CV evaluation, classifying eight emotions. The results revealed that these modalities carry relevant information to detect users’ emotional state and their combination enables improvement of system performance.