Recent technological developments have enabled computers to identify and categorize facial expressions to determine a person’s emotional state in an image or a video. This process, called “Facial Expression Recognition (FER)”, has become one of the most popular research areas in computer vision. In recent times, deep FER systems have primarily concentrated on addressing two significant challenges: the problem of overfitting due to limited training data availability, and the presence of expression-unrelated variations, including illumination, head pose, image resolution, and identity bias. In this paper, a comprehensive survey is provided on deep FER, encompassing algorithms and datasets that offer insights into these intrinsic problems. Initially, this paper presents a detailed timeline showcasing the evolution of methods and datasets in deep facial expression recognition (FER). This timeline illustrates the progression and development of the techniques and data resources used in FER. Then, a comprehensive review of FER methods is introduced, including the basic principles of FER (components such as preprocessing, feature extraction and classification, and methods, etc.) from the pro-deep learning era (traditional methods using handcrafted features, i.e., SVM and HOG, etc.) to the deep learning era. Moreover, a brief introduction is provided related to the benchmark datasets (there are two categories: controlled environments (lab) and uncontrolled environments (in the wild)) used to evaluate different FER methods and a comparison of different FER models. Existing deep neural networks and related training strategies designed for FER, based on static images and dynamic image sequences, are discussed. The remaining challenges and corresponding opportunities in FER and the future directions for designing robust deep FER systems are also pinpointed.

During the last years, Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance in image classification. Their architectures have largely drawn inspiration by models of the primate visual system. However, while recent research results of neuroscience prove the existence of non-linear operations in the response of complex visual cells, little effort has been devoted to extend the convolution technique to non-linear forms. Typical convolutional layers are linear systems, hence their expressiveness is limited. To overcome this, various non-linearities have been used as activation functions inside CNNs, while also many pooling strategies have been applied. We address the issue of developing a convolution method in the context of a computational model of the visual cortex, exploring quadratic forms through the Volterra kernels. Such forms, constituting a more rich function space, are used as approximations of the response profile of visual cells. Our proposed second-order convolution is tested on CIFAR-10 and CIFAR-100. We show that a network which combines linear and non-linear filters in its convolutional layers, can outperform networks that use standard linear filters with the same architecture, yielding results competitive with the state-of-the-art on these datasets.

The analysis of multimodal data collected by innovative imaging sensors, Internet of Things devices, and user interactions can provide smart and automatic distant monitoring of Parkinsons and Alzheimers patients and reveal valuable insights for early detection and/or prevention of events related to their health. This article describes a novel system that involves data capturing and multimodal fusion to extract relevant features, analyze data, and provide useful recommendations. The system gathers signals from diverse sources in health monitoring environments, understands the user behavior and context, and triggers proper actions for improving the patients quality of life. The system offers a multimodal, multi-patient, versatile approach not present in current developments. It also offers comparable or improved results for detection of abnormal behavior in daily motion. The system was implemented and tested during 10 weeks in real environments involving 18 patients.

In this work, a method for human fall detection is presented based on Recurrent Neural Networks. The ability of these networks to process and encode sequential data, such as acceleration measurements from body-worn sensors, makes them ideal candidates for this task. Furthermore, since such networks can benefit greatly from additional data during training, the use of a data augmentation procedure involving random 3D rotations has been investigated. When evaluated on the publicly available URFD dataset, the proposed method achieved better results compared to other methods.

Graph Neural Networks (GNNs) have become a powerful tool in order to learn from graph-structured data. Their ability to capture complex relationships and dependencies within graph structures, allows them to have a great number of applications in various domains, including social network analysis, recommendation systems and drug discovery. The aim of this work is to provide a detailed overview of the models in Graph Neural Networks and propose a new taxonomy of GNNs, including Deep Generative Models for graphs as a distinct category. The works included were selected based on their impact, with recent related papers also considered.