Video is a dominant traffic type in mobile networks. D2D communication, one of the most promising technologies for 6G, enables efficient localized distribution of content and thus reduces latency and bandwidth consumption for content delivery. Discovering potential mobile user devices and caching promotional content or short video content in D2D self-organizing communication-based social networks requires overcoming challenges such as heterogeneous structural features of user interactions and user state changes. We propose a graph neural network based GSVIA framework that integrates user information from the perspective of user interaction behavior and social features, and also propose an InfTMC algorithm for maximizing the weights of a set of caching nodes using aggregate function submodularity. Experimental results under two different weighted IC models show that the InfTMC algorithm improves the influence spread by an average of 46.80% and 45.44% compared to the existing algorithms.

In order to reduce the delay and bandwidth consumption of content delivery, we investigate how to uncover effective edge nodes in short video social network and cache promotional content or high-quality short video content in them. In this paper, we define the edge caching node selection problem for influence maximization under budget constraint, which proves to be an NP-hard problem. We propose an InfT algorithm based on influence spread tree to uncover edge nodes with high influence capacity in short video social network. During the selection of edge nodes, the cost changes dynamically with the selection of nodes. Experiments on scale-free network show that InfT makes the influence spread of edge caching nodes respectively improve by 21.82%, 438.21%, and 612.04%, when compared to some existing classical algorithms.

Existing methods of multi-person video 3D human Pose and Shape Estimation (PSE) typically adopt a two-stage strategy, which first detects human instances in each frame and then performs single-person PSE with temporal model. However, the global spatio-temporal context among spatial instances can not be captured. In this paper, we propose a new end-to-end multi-person 3D Pose and Shape estimation framework with progressive Video Transformer, termed PSVT. In PSVT, a spatio-temporal encoder (STE) captures the global feature dependencies among spatial objects. Then, spatio-temporal pose decoder (STPD) and shape decoder (STSD) capture the global dependencies between pose queries and feature tokens, shape queries and feature tokens, respectively. To handle the variances of objects as time proceeds, a novel scheme of progressive decoding is used to update pose and shape queries at each frame. Besides, we propose a novel pose-guided attention (PGA) for shape decoder to better predict shape parameters. The two components strengthen the decoder of PSVT to improve performance. Extensive experiments on the four datasets show that PSVT achieves stage-of-the-art results.

Multi-person 3D pose estimation is a challenging task because of occlusion and depth ambiguity, especially in the cases of crowd scenes. To solve these problems, most existing methods explore modeling body context cues by enhancing feature representation with graph neural networks or adding structural constraints. However, these methods are not robust for their single-root formulation that decoding 3D poses from a root node with a pre-defined graph. In this paper, we propose GR-M3D, which models the Multi-person 3D pose estimation with dynamic Graph Reasoning. The decoding graph in GR-M3D is predicted instead of pre-defined. In particular, It firstly generates several data maps and enhances them with a scale and depth aware refinement module (SDAR). Then multiple root keypoints and dense decoding paths for each person are estimated from these data maps. Based on them, dynamic decoding graphs are built by assigning path weights to the decoding paths, while the path weights are inferred from those enhanced data maps. And this process is named dynamic graph reasoning (DGR). Finally, the 3D poses are decoded according to dynamic decoding graphs for each detected person. GR-M3D can adjust the structure of the decoding graph implicitly by adopting soft path weights according to input data, which makes the decoding graphs be adaptive to different input persons to the best extent and more capable of handling occlusion and depth ambiguity than previous methods. We empirically show that the proposed bottom-up approach even outperforms top-down methods and achieves state-of-the-art results on three 3D pose datasets.

Video 3D human pose estimation aims to localize the 3D coordinates of human joints from videos. Recent transformer-based approaches focus on capturing the spatiotemporal information from sequential 2D poses, which cannot model the contextual depth feature effectively since the visual depth features are lost in the step of 2D pose estimation. In this paper, we simplify the paradigm into an end-to-end framework, Instance-guided Video Transformer (IVT), which enables learning spatiotemporal contextual depth information from visual features effectively and predicts 3D poses directly from video frames. In particular, we firstly formulate video frames as a series of instance-guided tokens and each token is in charge of predicting the 3D pose of a human instance. These tokens contain body structure information since they are extracted by the guidance of joint offsets from the human center to the corresponding body joints. Then, these tokens are sent into IVT for learning spatiotemporal contextual depth. In addition, we propose a cross-scale instance-guided attention mechanism to handle the variational scales among multiple persons. Finally, the 3D poses of each person are decoded from instance-guided tokens by coordinate regression. Experiments on three widely-used 3D pose estimation benchmarks show that the proposed IVT achieves state-of-the-art performances.

Path planning is a key technology to realize the automatic navigation of unmanned aerial vehicles (UAV), which has great significance both in theory and practical application. Evolutionary algorithms (EAs) are a type of nature-inspired computational methodologies for addressing complex real-world problems that cannot be solved well by mathematical or traditional modeling. However, the huge calculation of each iteration in the EAs greatly reduces the efficiency of the algorithm. In this paper, to accelerate the search speed of EAs and consider the dynamics of the environment, we propose a hierarchical recursive multi-agent genetic algorithm that can perform path planning in real time, termed as HR-MAGA. We used a strategy of layer-by-layer optimization on 3D maps with different resolution in the optimization process. The local search ability of the algorithm is improved by competition and self-learning process. In addition, the hierarchical recursive optimization process can greatly reduce the amount of computation and effectively deal with the dynamic characteristics of the environment. The experimental results show that HR-MAGA not only has strong global optimization ability, but more importantly, is able to generate collision free paths in real time after considering the physical limitations of the UAV and the dynamics of the environment.