Spatio-temporal action detection remains a challenging task due to the inherent coupling between precise spatial localization and effective temporal modeling. Although recent methods have achieved promising performance, they often suffer from spatial feature degradation during temporal aggregation and insufficient modeling of heterogeneous temporal dynamics across actions of varying durations. Moreover, most existing approaches fuse spatio-temporal information in a coarse manner, limiting their robustness in complex scenes. To address these challenges, we propose an efficient spatio-temporal perception network, termed STeP-Net. Specifically, we introduce a frequency-aware feature reconstruction module (FARM) to mitigate the loss of fine-grained spatial details caused by multi-scale temporal processing. Meanwhile, a directional spatial context module (DSCM) is designed to explicitly model local-global spatial dependencies for more robust spatial representation. Furthermore, we propose a heterogeneous temporal reasoning module (HTRM) to capture long-range temporal dependencies while accommodating diverse temporal patterns of different actions. Extensive experiments on three benchmark datasets, including AVA 2.2, UCF101-24, and the self-constructed AVA-E dataset, are conducted to evaluate the proposed method. STeP-Net achieves average mAPs of 18.2% on AVA 2.2 and 72.55% on AVA-E, and reaches 86.97% average mAP on UCF101-24 with a margin of 1.87% over the second-best result, indicating favorable performance across multiple benchmarks.

Tracking a moving unmanned ground vehicle (UGV) with an autonomous Unmanned Aerial Vehicle (UAV) is challenging, particularly in GNSS-denied indoor environments where reacquiring the UGV after losing track poses a significant obstacle. This paper presents a novel learning framework designed to address these challenges, enabling a quadrotor UAV to effectively chase a moving UGV and regain tracking in an indoor environment. The proposed framework encompasses two primary components: the Track-HCL and the Tracking Vision System (TVS). The TVS leverages a lightweight tracker to offer real-time recognition and localization of the UGV. Additionally, the Chronological Ghosting (CG) method is employed to describe the UGV’s motion trend within a single frame. The Track-HCL component introduces a hybrid curriculum strategy to guide policy learning for the Deep Reinforcement Learning (DRL) agent. The Track-HCL enables the agent to learn the tracking policy conducive to target chasing and proficient reacquisition. We demonstrate the effectiveness of the proposed method in both simulation and field experiments.

Abstract Online health communities (OHCs) function as significant platforms that people use to obtain information and emotional support. Despite many studies on user behavior and relationships, little attention has been paid to user identities and how different layers of identities are interwoven. To address this potential research gap, this study examined users' relationship‐layer identities and their evolution by elaborating on the communication theory of identity (CTI) and social support theory. Additionally, based on our previous study on users' personal‐layer identities in OHCs, we investigated how users' relationship‐layer identities interacted with their personal‐layer identities. This study classified users' posts and replies into providing informational support, seeking informational support, providing emotional support, seeking emotional support, and companionship using the bidirectional encoder representation from transformers (BERT), with F1‐scores above 0.848. Through social network analysis, this study found that users of OHCs constructed their relationship‐layer identities more through informational interactions than through emotional interactions. Users with various personal‐layer identities presented different relationship‐layer identities. Users' relationships were more initiated by information exchange, and users with more interactions had more companionship activities. OHCs provided efficient communication channels for people to exchange social support.

With the development of high-speed trains (HST), efficient and reliable communication services in high mobility scenarios have become an urgent demand. As one of the strong candidates in 5G wireless system, polar codes along with its optimized design should also be investigated under high mobility scenarios. In this paper, a scheme of hash-concatenated polar codes is proposed to reduce the false alarm rate, which is a key performance in 5G enhanced mobile broadband control channel. Then, for data channels, hash-based cyclic redundancy check (CRC)-aided polar codes with a joint successive cancellation list decoding method is introduced to improve the error-correcting performance. Simulation results show that the hash-concatenated polar codes can achieve both the lower false alarm rate and better error-correcting performance than conventional CRC-aided polar codes in both the AWGN and high mobility channels. Furthermore, with the joint decoding approach, hash-based CRC-aided polar codes perform better than LTE turbo codes for high-order modulations in terms of the frame error rate over the HST channel.