Radar-based human activity recognition (HAR) has gained significant attention due to its advantages in privacy protection, interference resistance, and low power consumption. However, existing systems face challenges such as slow model processing speed, limited practical applicability, and a strong dependence on large labeled datasets. A new radar-based HAR system designed for multi-scenario and multi-directional application is proposed in this paper. First, a small-sample human activity dataset (MSMD-RHD) is developed, capturing activities from multiple directions and scenarios to better simulate real-life conditions. Next, we propose a lightweight, multi-dimensional feature enhancement and fusion network, HogEffNet, which integrates EfficientNetB1 and Histogram of Oriented Gradients (HOG) to extract behavioral information from Time-Doppler (TD) maps. To mitigate overfitting caused by limited samples, we apply a staged transfer learning strategy to fine-tune the top parameters of the pre-trained EfficientNetB1. HOG is incorporated to address the insufficient ability of EfficientNetB1 in extracting low-dimensional local shape features, thereby improving robustness in multi-directional HAR. Furthermore, Sparse Autoencoder (SAE) is employed to fuse high-dimensional and low-dimensional features effectively. The proposed system is evaluated on two datasets. Ablation experiments conducted on the MSMD-RHD dataset achieve an accuracy of 98.8%, demonstrating the effectiveness of each component. Comparative experiments on the public dataset of University of Glasgow, UK, achieve an accuracy of 97.2%, surpassing recent state-of-the-art methods.

Relationships in online social networks often imply social connections in real life. An accurate understanding of relationship types benefits many applications, e.g. social advertising and recommendation. Some recent attempts have been proposed to classify user relationships into predefined types with the help of pre-labeled relationships or abundant interaction features on relationships. Unfortunately, both relationship feature data and label data are very sparse in real social platforms like WeChat, rendering existing methods inapplicable. In this paper, we present an in-depth analysis of WeChat relationships to identify the major challenges for the relationship classification task. To tackle the challenges, we propose a Local Community-based Edge Classification (LoCEC) framework that classifies user relationships in a social network into real-world social connection types. LoCEC enforces a three-phase processing, namely local community detection, community classification and relationship classification, to address the sparsity issue of relationship features and relationship labels. Moreover, LoCEC is designed to handle large-scale networks by allowing parallel and distributed processing. We conduct extensive experiments on the real-world WeChat network with hundreds of billions of edges to validate the effectiveness and efficiency of LoCEC.

Gatekeeping is one of the most important tasks in MSL, especially defending aerial shots. This paper proposes a goalkeeper strategy based on combining two image sources. The goalkeeper uses the Omni-vision system to adjust heading orientation of the Kinect sensor for capturing both depth and RGB images of the ball. The depth information is used to detect the spatial position of the ball and the RGB image ensures the validity of the recognition. Once opponents shoot, the proposed system uses the first few detected positions of the ball to predict the interception point by the goalkeeper based on the least square method. The accuracy and effectiveness of the proposed strategy are verified by experimentation.