Fuyuan Liu, Dianyu Yu, He Ren, Nayu Liu, Xiaomian Kang, Delai Qiu, Fa Zhang, Genpeng Zhen, Shengping Liu, Liang Jiaen, Weihuang, Yining Wang, Junnan Zhu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

In vehicle-road cooperation, the advancement of vehicle-side autonomous driving is hindered by bottlenecks, such as limited sensing range, computational power, and environmental adaptability. Collaboration with roadside units is essential for achieving more accurate and complex spatial perception. This study presents a new prototype to enhancing spatial perception in road environments within Augmented Intelligence of Things (AIoT) systems using light detection and ranging (LiDAR) technology. Unlike traditional AIoT systems, which rely on cameras and struggle in complex conditions, the proposed prototype uses an edge-cloud–terminal collaborative sensing model to enhance 3-D spatial perception. A notable feature of this prototype is the integration of the distance and density adaptive filtering (DDAF) method, which ensures efficient point cloud filtering at the edge, with an average F1-score of 96.03% and an average latency of 11.78 ms, demonstrating strong accuracy and low latency across various scenarios. The incorporation of DDAF further improves the mean average precision (mAP) of deep learning-based 3-D object detection on the cloud by 2.34%, reduces processing time by 83.54%, and decreases peak memory usage by 90.18%, facilitating precise 3-D spatial analysis. The final results are displayed in real-time on the terminal for visualization and interaction. The efficacy of this prototype is demonstrated through a real-world case study. This research highlights the role of LiDAR and AIoT in overcoming spatial perception challenges in vehicle-road cooperation, leading to safer, more efficient transportation solutions.

Infrared small target detection (IRSTD) has been widely applied and developed in military and civilian fields, playing a vital role. Despite the extensive research foundation of traditional manual feature-based methods, they are still constrained by the inherent problem of infrared small targets lacking prior features. In recent years, the advancement of deep learning methods has enriched the research landscape in this field, yet they are still constrained by the imbalance of positive and negative samples between the target and the background. To address these issues, we propose a novel prior guided dense nested network (PGDN-Net), which ingeniously integrates traditional manual features with a deep learning network model. First, three prior features are extracted, including the high-order Riesz transform feature, the compactness and heterogeneity feature (CH), and the corner feature of the structure tensor (ST). Then, these features are input into a dense nested network for guidance, supported by a two-orientation attention aggregation module and a channel and spatial attention module. Different features play their respective guiding roles in different depths of the network. Through multiple attention mechanisms and feature fusion operations on the interested target area, the extraction and preservation of target features can be improved, while easily removing irrelevant backgrounds. Experiments on public datasets demonstrate the effectiveness and progressiveness of our PGDN-Net. Compared with other state-of-the-art methods, it achieves better performance in background suppression, target enhancement, probability of detection, and false alarm rate. In addition, the PGDN-Net model can effectively maintain and restore the original shape of the target while performing robust detection, which is beneficial for subsequent fine-grained recognition tasks.

Recent deep learning models for 3-D object detection (3DOD) based on LiDAR point clouds have exhibited promising performance in the detection of on-road objects. However, their performance is often constrained when detecting small objects like pedestrians and cyclists. Due to occlusions and their inherent volumes, the spatial and semantic information of small objects is relatively limited. Therefore, we propose a point distribution awareness network (PDANet) that emphasizes the structural distribution of objects, to enhance the extraction of positional, geometric, and semantic features. It comprises two modules: a multifeature adaptive perception (MFAP) module, enhancing local feature perception, and a semantic-aware downsampling module, enhancing the selection of foreground points. These modules can be used in a plug-and-play manner for point downsampling feature perception. They are then integrated as a branch into representative point-based (IA-SSD), point-voxel-based (PV-RCNN), and voxel-based (Centerpoint) detectors, resulting in PDA-SSD, PDA-PV-RCNN, and PDA-CenterPoint models. Through experiments on the ONCE dataset, the PDANet brought up the mean average precision (mAP) of all three types of baseline models by 3.03%, 4.60%, and 6.15%, respectively. Further experiments on the KITTI dataset illustrate an improvement in average precision (AP) of small objects up to 7.41%.