Efficient Crowd Counting via Dual Knowledge Distillation.

Most researchers focus on designing accurate crowd counting models with heavy parameters and computations but ignore the resource burden during the model deployment. A real-world scenario demands an efficient counting model with low-latency and high-performance. Knowledge distillation provides an elegant way to transfer knowledge from a complicated teacher model to a compact student model while maintaining accuracy. However, the student model receives the wrong guidance with the supervision of the teacher model due to the inaccurate information understood by the teacher in some cases. In this paper, we propose a dual-knowledge distillation (DKD) framework, which aims to reduce the side effects of the teacher model and transfer hierarchical knowledge to obtain a more efficient counting model. First, the student model is initialized with global information transferred by the teacher model via adaptive perspectives. Then, the self-knowledge distillation forces the student model to learn the knowledge by itself, based on intermediate feature maps and target map. Specifically, the optimal transport distance is utilized to measure the difference of feature maps between the teacher and the student to perform the distribution alignment of the counting area. Extensive experiments are conducted on four challenging datasets, demonstrating the superiority of DKD. When there are only approximately 6% of the parameters and computations from the original models, the student model achieves a faster and more accurate counting performance as the teacher model even surpasses it.

Fabric computing: Concepts, opportunities, and challenges.

With the advent of the Internet of Everything, people can easily interact with their environments immersively. The idea of pervasive computing is becoming a reality, but due to the inconvenience of carrying silicon-based entities and a lack of fine-grained sensing capabilities for human-computer interaction, it is difficult to ensure comfort, esthetics, and privacy in smart spaces. Motivated by the rapid developments in intelligent fabric technology in the post-Moore era, we propose a novel computing approach that creates a paradigm shift driven by fabric computing and advocate a new concept of non-chip sensing in living spaces. We discuss the core notion and benefits of fabric computing, including its implementation, challenges, and future research opportunities.

Depression Analysis and Recognition Based on Functional Near-Infrared Spectroscopy.

Depression is the result of a complex interaction of social, psychological and physiological elements. Research into the brain disorders of patients suffering from depression can help doctors to understand the pathogenesis of depression and facilitate its diagnosis and treatment. Functional near-infrared spectroscopy (fNIRS) is a non-invasive approach to the detection of brain functions and activities. In this paper, a comprehensive fNIRS-based depression-processing architecture, including the layers of source, feature and model, is first established to guide the deep modeling for fNIRS. In view of the complexity of depression, we propose a methodology in the time and frequency domains for feature extraction and deep neural networks for depression recognition combined with current research. It is found that compared to non-depression people, patients with depression have a weaker encephalic area connectivity and lower level of activation in the prefrontal lobe during brain activity. Finally, based on raw data, manual features and channel correlations, the AlexNet model shows the best performance, especially in terms of the correlation features and presents an accuracy rate of 0.90 and a precision rate of 0.91, which is higher than ResNet18 and machine-learning algorithms on other data. Therefore, the correlation of brain regions can effectively recognize depression (from cases of non-depression), making it significant for the recognition of brain functions in the clinical diagnosis and treatment of depression.

Intelligent Traffic Adaptive Resource Allocation for Edge Computing-based 5G Networks.

The popularity of smart mobile devices has led to a tremendous increase in mobile traffic, which has put a considerable strain on the fifth generation of mobile communication networks (5G). Among the three application scenarios covered by 5G, ultra-high reliability and ultra-low latency (uRLLC) communication can best be realized with the assistance of artificial intelligence. For a combined 5G, edge computing and IoT-Cloud (a platform that integrates the Internet of Things and cloud) in particular, there remains many challenges to meet the uRLLC latency and reliability requirements despite a tremendous effort to develop smart data-driven methods. Therefore, this paper mainly focuses on artificial intelligence for controlling mobile-traffic flow. In our approach, we first develop a traffic-flow prediction algorithm that is based on long short-term memory (LSTM) with an attention mechanism to train mobile-traffic data in single-site mode. The algorithm is capable of effectively predicting the peak value of the traffic flow. For a multi-site case, we present an intelligent IoT-based mobile traffic prediction-and-control architecture capable of dynamically dispatching communication and computing resources. In our experiments, we demonstrate the effectiveness of the proposed scheme in reducing communication latency and its impact on lowering packet-loss ratio. Finally, we present future work and discuss some of the open issues.

Wireless Fractal Ultra-Dense Cellular Networks.

With the ever-growing number of mobile devices, there is an explosive expansion in mobile data services. This represents a challenge for the traditional cellular network architecture to cope with the massive wireless traffic generated by mobile media applications. To meet this challenge, research is currently focused on the introduction of a small cell base station (BS) due to its low transmit power consumption and flexibility of deployment. However, due to a complex deployment environment and low transmit power of small cell BSs, the coverage boundary of small cell BSs will not have a traditional regular shape. Therefore, in this paper, we discuss the coverage boundary of an ultra-dense small cell network and give its main features: aeolotropy of path loss fading and fractal coverage boundary. Simple performance analysis is given, including coverage probability and transmission rate, etc., based on stochastic geometry theory and fractal theory. Finally, we present an application scene and discuss challenges in the ultra-dense small cell network.

Mobility-Aware Caching and Computation Offloading in 5G Ultra-Dense Cellular Networks.

Recent trends show that Internet traffic is increasingly dominated by content, which is accompanied by the exponential growth of traffic. To cope with this phenomena, network caching is introduced to utilize the storage capacity of diverse network devices. In this paper, we first summarize four basic caching placement strategies, i.e., local caching, Device-to-Device (D2D) caching, Small cell Base Station (SBS) caching and Macrocell Base Station (MBS) caching. However, studies show that so far, much of the research has ignored the impact of user mobility. Therefore, taking the effect of the user mobility into consideration, we proposes a joint mobility-aware caching and SBS density placement scheme (MS caching). In addition, differences and relationships between caching and computation offloading are discussed. We present a design of a hybrid computation offloading and support it with experimental results, which demonstrate improved performance in terms of energy cost. Finally, we discuss the design of an incentive mechanism by considering network dynamics, differentiated user's quality of experience (QoE) and the heterogeneity of mobile terminals in terms of caching and computing capabilities.

The impact of color combinations on the legibility of text presented on LCDs.

The paper investigates the impact of text and background color combinations on the legibility of text presented on LCDs. The legibility of 56 elementary color combinations was tested by 308 participants. The first task required the participants to identify 21 uppercase alphabetic characters selected and presented in conformance with a Snellen chart in various color combinations. For each color combination, the number of correctly identified characters was recorded as a visual performance measure. In the second task, participants subjectively rated the legibility of color combinations on a ten-point Likert scale. The results show that neither the Le Courier legibility table nor the CRT legibility table is appropriate for LCDs. A new legibility table is proposed for LCDs where the highest ranked are contrastive color combinations with positive polarity whereas for CRT displays, the highest ranked are contrastive color combinations with negative polarity. The findings of this study can be used to determine the best possible color combinations when developing content displayed on LCDs.