As a potential application of Vehicle-to-Everything (V2X) communication, multi-agent collaborative perception has achieved significant success in 3D object detection. While these methods have demonstrated impressive results on standard benchmarks, the robustness of such approaches in the face of complex real-world environments requires additional verification. To bridge this gap, we introduce the first comprehensive benchmark designed to evaluate the robustness of collaborative perception methods in the presence of natural corruptions typical of real-world environments. Furthermore, we propose DSRC, a robustness-enhanced collaborative perception method aiming to learn Density-insensitive and Semantic-aware collaborative Representation against Corruptions. DSRC consists of two key designs: i) a semantic-guided sparse-to-dense distillation framework, which constructs multi-view dense objects painted by ground truth bounding boxes to effectively learn density-insensitive and semantic-aware collaborative representation; ii) a feature-to-point cloud reconstruction approach to better fuse critical collaborative representation across agents. To thoroughly evaluate DSRC, we conduct extensive experiments on real-world and simulated datasets. The results demonstrate that our method outperforms state-of-the-art collaborative perception methods in both clean and corrupted conditions.

With the emergence of the Internet of Things (IoT), indoor positioning services have garnered increasing attention. When it comes to target tracking, the uncertainty surrounding velocity poses a challenge in obtaining the initial state, and the setting of this initial state significantly impacts early tracking performance. To address this issue, this paper proposes an extended Kalman filter (EKF) algorithm based on a dual model, aiming to mitigate the influence of this problem. Although the tracking performance of the without velocity constrained (WVC)-EKF model falls short compared to that of the velocity constraint (VC)-EKF model, the WVC-EKF model eliminates the need to grapple with the difficulty of setting the initial velocity. Consequently, the tracking process benefits from the combined strengths of both models. Firstly, the early tracking is accomplished using the WVC-EKF model. Subsequently, the WVC-EKF model serves as an auxiliary component to assist the VC-EKF model in establishing the initial state. Finally, the VC-EKF model takes charge of the subsequent tracking. In this paper, ray-tracing software was employed for simulation purposes, and the obtained results demonstrate that the proposed algorithm enhances tracking accuracy.

While exploiting body segmentations for supervision, existing 3D human pose and shape estimation methods are plagued by mismatches between clothed body segmentations and skinned SMPL model reprojections. Moreover, noisy pixels introduced by inaccurate segmentation annotations also prevent the model from improving the reconstruction performance further. To address these problems, we propose a novel generalizable framework called Uncertainty-aware body Part Segmentation (UPS), which penalizes different body parts with data uncertainty estimation. Specifically, we use sample-specific segmentations to supervise skinned and clothed body parts separately for realistic human mesh recovery. Furthermore, we leverage data uncertainty to improve the model capacity via learning from representative pixels and resisting noisy ones. Our extensive qualitative and quantitative experiments show that UPS achieves competitive reconstruction results on standard benchmarks.

Edge computing has emerged as a promising paradigm for addressing the challenges of latency, bandwidth, and energy consumption in the era of big data and the intelligent Internet of Things. However, the limited computing resources of edge devices and their vulnerability to failures pose significant challenges to the reliability and availability of edge computing systems. To this end, we propose a novel architecture for reliable edge computing that leverages the collective computing power of mobile edge devices in this article. Our architecture employs a task-oriented triple-stage monitoring mechanism to ensure system reliability. Moreover, we present a shared computing framework that allows edge devices to dynamically share their computing resources based on the current availability and workload. We evaluate the effectiveness of the proposed architecture with several computational tasks, including$\pi $calculation and video processing. The results show that our architecture achieves high reliability and availability while also improving the performance and energy efficiency of edge devices.