Recently, a prevailing trend of user generated content (UGC) on social media sites is the emerging micro-videos. Micro-videos afford many potential opportunities ranging from network content caching to online advertising, yet there are still little efforts dedicated to research on micro-video understanding. In this paper, we focus on popularity prediction of micro-videos by presenting a novel low-rank multi-view embedding learning framework. We name it as transductive low-rank multi-view regression (TLRMVR), and it is capable of boosting the performance of micro-video popularity prediction by jointly considering the intrinsic representations of the source and target samples. In particular, TLRMVR integrates low-rank multi-view embedding and regression analysis into a unified framework such that the lowest-rank representation shared by all views not only captures the global structure of all views, but also indicates the regression requirements. The framework is formulated as a regression model and it seeks a set of view-specific projection matrices with low-rank constraints to map multi-view features into a common subspace. In addition, a graph regularization term is constructed to improve the generalisation capability and further prevents the overfitting problem. Extensive experiments conducted on a publicly available dataset demonstrate that our proposed method achieve promising results as compared with state-of-the-art baselines.

With the rapid development of the wireless sensor network and the continuous improvement of its key technologies, the concept of Internet of Things (IoT) has been encouraged and extended due to its wide applications in scenarios such as smart homes and healthcare. Under the background, human activity recognition (HAR) has drawn great attention in recent years. In this paper, we present a discriminant approach to recognize daily human activities recorded through accelerometer sensor. In the proposed approach, we first use S transform (ST) to extract features, and then introduce a supervised regularization-based robust subspace (SRRS) learning method to learn low-dimensional intrinsic feature representation from the original feature subspace. Particularly, ST has been described as a joint time-frequency representation (TFR), which is insensitive to noise. SRRS can learn more robust and discriminative features to reinforce the descriptions of samples while removing noise and redundancy. Experiments are conducted on three publicly available datasets i.e., WISDM, SCUT-NAA and mHealth demonstrating the superior performance of our proposed scheme compared with state-of-the-art methods.

Eye-tracking technology has become a powerful tool for biomedical-related applications due to its simplicity of operation and low requirements on patient language skills. This study aims to use the machine-learning models and deep-learning networks to identify key features of eye movements in Alzheimer's Disease (AD) under specific visual tasks, thereby facilitating computer-aided diagnosis of AD. Firstly, a three-dimensional (3D) visuospatial memory task is designed to provide participants with visual stimuli while their eye-movement data are recorded and used to build an eye-tracking dataset. Then, we propose a novel deep-learning-based model for identifying patients with Alzheimer's Disease (PwAD) and healthy controls (HCs) based on the collected eye-movement data. The proposed model utilizes a nested autoencoder network to extract the eye-movement features from the generated fixation heatmaps and a weight adaptive network layer for the feature fusion, which can preserve as much useful information as possible for the final binary classification. To fully verify the performance of the proposed model, we also design two types of models based on traditional machine-learning and typical deep-learning for comparison. Furthermore, we have also done ablation experiments to verify the effectiveness of each module of the proposed network. Finally, these models are evaluated by four-fold cross-validation on the built eye-tracking dataset. The proposed model shows 85% average accuracy in AD recognition, outperforming machine-learning methods and other typical deep-learning networks.

Time-series prediction has become a prominent challenge, especially when the data are described as sequences of multiway arrays. Because noise and redundancy may exist in the tensor representation of a time series, we focus on solving the problem of high-order time-series prediction under a tensor decomposition framework and develop two novel multilinear models: 1) the multilinear orthogonal autoregressive (MOAR) model and 2) the multilinear constrained autoregressive (MCAR) model. The MOAR model is designed to preserve as much information as possible from the original tensorial data under orthogonal constraints. The MCAR model is an enhanced version that is developed by replacing orthogonal constraints with an inverse decomposition error term. For both models, we project the original tensor into subspaces spanned by basis matrices to facilitate the discovery of the intrinsic temporal structure embedded in the original tensor. To build connections among consecutive slices of the tensor, we generalize a traditional autoregressive model to tensor form to better preserve the temporal smoothness. Experiments conducted on four publicly available datasets demonstrate that our proposed methods converge within a small number of iterations during the training stage and achieve promising results compared with state-of-the-art methods.

There is a growing demand for monitors to provide high quality visualization with more bits representing each rendered pixel. However, since most existing images and videos are of low bit-depth (LBD), transforming LBD images to visually pleasant high bit-depth (HBD) versions is of significant value. Most existing bit-depth enhancement methods generate unsatisfactory HBD images with annoying false contour artifacts or blurry details, and some algorithms are also time-consuming. To overcome these drawbacks, we propose a bit-depth enhancement framework via concatenating all level features of deep neural networks (DNN). A novel deep learning network is proposed based on deep convolutional variational auto-encoders (VAE), and skip connections that concatenate every two layers are applied to pass low-level and high-level features to consequent layers, easing the gradient vanishing problem. Meanwhile, the proposed network is optimized to generate the residual between original images and its quantized ones, which performs better than recovering HBD images directly. Experimental results show the proposed algorithm can eliminate false contour artifacts of the recovered HBD images with low time consumption, and can achieve dramatic restoration performance gains compared with state-of-the-art methods both subjectively and objectively.