In skeleton-based action recognition, the approaches based on graph convolutional networks(GCN) have achieved remarkable performance by modeling spatial-temporal graphs to explore the physical dependencies between body joints. However, these methods mostly apply hierarchical GCNs to aggregate wider-range neighborhood information, which makes joint features be weakened during long diffusion. In this paper, we design a multi-scale mixed dense graph convolutional network (MMDGCN) to overcome both shortcomings. We propose a dense graph convolution operation to enhance the local context information of joints, and then the spatial and temporal attention modules with a larger receptive field are introduced to help the model strengthen the discriminative features to adaptively refine the intermediate feature maps. We also design a multi-scale mixed temporal convolution module, which provides a flexible temporal graph through the combination of different scale convolution kernels. Extensive experiments on the three real-world datasets (NTU-RGB+D, NTU-RGB+D120 and Kinetics) demonstrate that the performance of the proposed MMDGCN in skeleton-based action recognition.

This article proposes a hybrid dimming (HD) scheme based on joint light-emitting diode (LED) selection and precoding design (TASP-HD) for multiple-user (MU) multiple-cell (MC) visible light communications (VLCs) systems. In TASP-HD, both the LED selection and the precoding of each cell can be dynamically adjusted to reduce the intra- and inter-cell interferences while satisfying illumination constraints. First, an MU-MC-VLC system model is established, and then a sum-rate maximization problem under the dimming level and illumination uniformity constraints is formulated. In this studied problem, the indices of activated LEDs and precoding matrices are optimized, which result in a complex nonconvex mixed-integer problem. To solve this problem, the original problem is separated into two subproblems. The first subproblem, which maximizes the sum rate of users via optimizing the LED selection with a given precoding matrix, is a mixed-integer problem solved by the penalty method. With the optimized LED selection matrix, the second subproblem which focuses on the maximization of the sum-rate via optimizing the precoding matrix is solved by the Lagrangian dual method. Finally, these two subproblems are iteratively solved to obtain a convergent solution. Simulation results verify that in a typical indoor scenario under a dimming level of 70%, the mean bandwidth efficiency (MBE) of TASP-HD is 4.8 bit/s/Hz and 7.13 bit/s/Hz greater than analog dimming (AD) and digital dimming (DD), respectively.

Tracking detailed hand motion is a fundamental research topic in the area of human-computer interaction (HCI) and has been widely studied for decades. Existing solutions with single-model inputs either require tedious calibration, are expensive or lack sufficient robustness and accuracy due to occlusions. In this study, we present a real-time system to reconstruct the exact hand motion by iteratively fitting a triangular mesh model to the absolute measurement of hand from a depth camera under the robust restriction of a simple data glove. We redefine and simplify the function of the data glove to lighten its limitations, i.e., tedious calibration, cumbersome equipment, and hampering movement and keep our system lightweight. For accurate hand tracking, we introduce a new set of degrees of freedom (DoFs), a shape adjustment term for personalizing the triangular mesh model, and an adaptive collision term to prevent self-intersection. For efficiency, we extract a strong pose-space prior to the data glove to narrow the pose searching space. We also present a simplified approach for computing tracking correspondences without the loss of accuracy to reduce computation cost. Quantitative experiments show the comparable or increased accuracy of our system over the state-of-the-art with about 40% improvement in robustness. Besides, our system runs independent of Graphic Processing Unit (GPU) and reaches 40 frames per second (FPS) at about 25% Central Processing Unit (CPU) usage.

In this study, the authors focus on energy efficiency (EE) optimisation via the cooperation of small cell base stations (SBSs) in dense small cell networks (DSCNs) where the control plane (C‐plane) and data plane (D‐plane) are decoupled. They propose a coalition‐based sleep mode and power allocation (CSMPA) scheme to improve the DSCN EE while guaranteeing the target rates of users and maintaining the system capacity. In the CSMPA scheme, the cooperation of SBSs is formulated as a coalitional game in partition form and a centralised heuristic coalition formation algorithm without power cost is developed to achieve the final stable coalition structure. Each SBS can serve users in the active subframes and sleep in those subframes without data transmission. Afterwards, since the interference between coalitions affects EE, a distributed price‐based power allocation algorithm is presented to optimise the transmit power of active SBSs per subframe. System‐level simulation results show that the proposed CSMPA scheme can yield less number of outage users and significant EE performance gain without jeopardising system capacity.

3D human pose estimation from a monocular RGB image is a challenging task in computer vision because of depth ambiguity in a single RGB image. As most methods consider joint locations independently which can lead to an overfitting problem on specific datasets, it’s crucial to consider the plausibility of 3D poses in terms of their overall structures. In this paper, we present Generative Adversarial Networks (GANs) for 3D human pose estimation, which learn plausible 3D human body representations by adversarial training. In GANs, the generator regresses 3D joint positions from a 2D input and the discriminator aims to distinguish the ground-truth 3D samples from the predicted ones. We leverage Graph Convolutional Networks (GCNs) in both generator and discriminator to fully exploit the spatial relations of input and output coordinates. The combination of GANs and GCNs promotes the network to predict more accurate 3D joint locations and learn more reasonable human body structures at the same time. We demonstrate the effectiveness of our approach on standard benchmark datasets (i.e. Human3.6M and HumanEva-I ) where it outperforms state-of-the-art methods. Furthermore, we propose a new evaluation metric distance-based Pose Structure Score (dPSS) for evaluating the structural similarity degree between the predicted 3D pose and its ground-truth.

Energy efficiency (EE) has become a crucial metric in dense small cell networks (DSCNs) due to the tremendous escalation of energy consumption. Since small cell base stations (SBSs) are randomly deployed, loads of them are usually different. In this letter, we aim to maximize the load-aware weighted sum of EEs (LWSEEs) across all subchannels in DSCNs jointly considering SBSs' loads and users' signal processing power through power allocation. First, we design the EE weight as a function of load factor (LF). Then, the LWSEE optimization problem is formulated as a non-concave sum-of-ratios optimization and we present the optimal LWSEE algorithm to tackle the problem based on the concave-convex procedure method. Furthermore, a simplified LWSEE (LWSEE-simp) algorithm where SBSs form interference groups is developed to reduce the signaling overhead in DSCNs. Simulation results show that SBSs' sum-EEs can be optimized according to their LFs and the optimal interference group size exists in the LWSEE-simp algorithm.