Person re-identification (Re-ID) is a fundamental service for AI-driven human centric computing and smart cities. Occlusion is one of the major challenges for the practical use of ReID. Existing work either use compute-intensive models or enhanced dataset to deal with occlusion. However, the single-model-driven works fail to find a reasonable balance between ReID accuracy and efficiency. To address this, we propose a Structured Occlusion Perception (SOP) framework based on multi-model collaboration. SOP first determines the occlusion level and then dynamically switches between lightweight and compute-intensive Re-ID models for Re-ID with light/heavy occlusion. SOP's occlusion decision extracts skeleton keypoints and calculates two weighted factors (spatial interaction and body part visibility) to distinguish occlusion levels, incurring little extra overhead. This design effectively reduces redundant computation while preserving recognition performance. Experiments on the DukeMTMC and Occluded-Duke datasets demonstrate that SOP reduces mAP by only 1.2 % and 1.5 %, respectively, while achieving computational saving of approximately 38.9 % and$\mathbf{4 1. 2 \%} \boldsymbol{.}$Furthermore, on the occlusion-labeled Occ-ReID dataset, SOP attains 84.8 % occlusion detection accuracy, demonstrating its effectiveness and robustness in cross-model inference and occlusion perception.

Software-Defined Vehicular Network (SDVN) is a networking architecture that can provide centralized control for vehicular networks. However, the design for routing policies in SDVNs is generally influenced by several limitations, such as frequent topological changes, complex service requests, and long model training time. Intelligent Digital Twin-based Software-Defined Vehicular Networks (IDT-SDVN) can overcome these weaknesses and maximize the advantages of the conventional SDVN architecture by enabling the controller to construct virtual network spaces and provide virtual instances of corresponding physical objects within the Digital Twin (DT). In this paper, we propose a junction-based hierarchical routing scheme in IDT-SDVN, namely, intelligent digital twin hierarchical (ELITE) routing. The proposed scheme is conducted in four phases: policy training and generation in the virtual network, and deployment and relay selection in physical networks. First, the policy learning phase employs several parallel agents in DT networks and derives multiple single-target policies. Second, the generation phase combines the learned policies and generates new policies based on complex communication requirements. Third, the deployment phase selects the most suitable generated policy according to the real-time network status and message types. A road path is calculated by the controller based on the selected policy and then sent to the requester vehicle. Finally, the relay selection phase is utilized to determine relay vehicles in a hop-by-hop process along the selected path. Simulation results demonstrate that ELITE achieves substantial improvements in terms of packet delivery ratio, end-to-end delay, and communication overhead compared with its counterparts.