arXiv subjects

Tianheng Zhu

Publications and source records attributed to Tianheng Zhu.

6 recordsLinked to original sources

arXiv2026

A Systematic Evaluation of Infrastructure-Based Radar System for Highway Traffic Monitoring

Infrastructure-based radar systems offer robust and long-range solutions for traffic monitoring, yet their detection and tracking performance under real-world conditions remains insufficiently evaluated. This study introduces DRaT (Drone and Radar Trajectories), a dual-modality dataset of naturalistic vehicle trajectories collected at a highway merging segment in Fort Worth, Texas, to systematically assess radar sensing performance against drone-derived ground truth. The performance is evaluated at three levels: individual vehicle detection, trajectory tracking, and macroscopic traffic parameter estimation. For individual vehicle detection, the radar achieves an overall precision of 78% and a recall of 57%, with degraded performance under congested traffic conditions and at longer distances. At the trajectory level, the radar demonstrates reasonably strong tracking performance (IDF1 = 0.699), maintaining reliable vehicle identities when tracks are successfully established. For macroscopic traffic flow metrics, the radar accurately estimates space-mean speed (MAPE < 4%) but underestimates density and volume by approximately 23% due to missed detections. The paper also discusses practical deployment considerations and potential downstream applications of roadside radar sensing systems. To support reproducible research on infrastructure-based sensing systems, we have open-sourced the DRaT dataset on Zenodo: https://zenodo.org/records/20171110.

cs.CV↗

arXiv2026

Drones as Annotators: Amodal 3D Auto-Labeling for Ground LiDAR with Aerial Priors

Scaling perception data in autonomous driving is hindered by manual 3D bounding box annotation, a costly and labor intensive process requiring substantial domain expertise. Existing auto-labeling methods reduce this burden, but most of them rely on onboard sensors, where a single ground-level viewpoint yields occluded and sparse observations and inaccurate object geometry. We introduce DAA (Drones as Annotators), a drone-assisted training-free framework for amodal 3D auto-labeling. DAA augments ground LiDAR with an aerial agent that provides less occluded vehicle detections with continuous tracks. Ground LiDAR, in turn, provides precise metric 3D geometry unavailable from aerial imagery alone. DAA exploits the two complementary yet heterogeneous modalities through a two-stage framework: (i) air-ground coordinate alignment unifies aerial detections and ground LiDAR scans into a shared map frame and bootstraps coarse 3D bounding boxes from the aligned aerial detections; (ii) EM-like amodal box refinement alternates between identifying the vehicle's foreground points and updating the coarse box from the identified foreground and aerial priors. We evaluate DAA on an air-ground cooperative perception dataset, where it consistently outperforms existing auto-labeling baselines. LiDAR detectors trained on DAA-generated labels remain competitive with those trained on manual annotations at moderate IoU thresholds. Evaluation against a reference vehicle with known poses and dimensions further confirms its accuracy. We also demonstrate that DAA transfers, without parameter retuning, from onboard to stationary roadside LiDAR and to multi-agent fused point clouds. Code and dataset will be released upon publication.

eess.IV↗

arXiv2026

Empowering Microscopic Traffic Simulators with Realistic Perception using Surrogate Sensor Models

Simulation is central to the evaluation of intelligent transportation system (ITS) applications. As ITS increasingly incorporates autonomous vehicle (AV) technologies as fleet vehicles and/or mobile sensors, accurate modeling of their perception capabilities becomes essential in high-fidelity simulations. While game-engine-based simulators reproduce realistic perception environments through 3D scene rendering and raw sensor data generation, they face scalability challenges in simulating traffic networks with a large number of AVs due to high computational cost. In contrast, microscopic traffic simulators (MTS) can scale efficiently but lack perception modeling capabilities. To bridge this gap, we propose MIDAR, a surrogate LiDAR detection model that mimics realistic LiDAR detections using only high-level features readily available from MTS. Specifically, MIDAR predicts true-positive and false-negative LiDAR detections based on the relative positions and dimensions of surrounding objects. To capture LiDAR visibility and occlusion effects, MIDAR introduces a ray-hit feature and a Refined Multi-hop Line-of-Sight (RM-LoS) graph processed by a geometry-aware Graph Transformer. MIDAR achieves an AUC of 0.94 in approximating LiDAR detection results using CARLA-generated point cloud data, and an AUC of 0.86 with real-world data from the nuScenes dataset. Two ITS applications, cooperative-perception-based adaptive signal control and vehicle trajectory reconstruction, are integrated with MIDAR to further validate its realism and necessity. Results show that MIDAR generates more realistic detection outputs as well as application-level performance metrics than simplified perception models while introducing minimal computational overhead, enabling seamless integration into large-scale, real-time traffic simulations. The code and data are publicly available at https://github.com/Purdue-CART-Lab/MIDAR.

cs.CV↗

arXiv2025

EGSTalker: Real-Time Audio-Driven Talking Head Generation with Efficient Gaussian Deformation

This paper presents EGSTalker, a real-time audio-driven talking head generation framework based on 3D Gaussian Splatting (3DGS). Designed to enhance both speed and visual fidelity, EGSTalker requires only 3-5 minutes of training video to synthesize high-quality facial animations. The framework comprises two key stages: static Gaussian initialization and audio-driven deformation. In the first stage, a multi-resolution hash triplane and a Kolmogorov-Arnold Network (KAN) are used to extract spatial features and construct a compact 3D Gaussian representation. In the second stage, we propose an Efficient Spatial-Audio Attention (ESAA) module to fuse audio and spatial cues, while KAN predicts the corresponding Gaussian deformations. Extensive experiments demonstrate that EGSTalker achieves rendering quality and lip-sync accuracy comparable to state-of-the-art methods, while significantly outperforming them in inference speed. These results highlight EGSTalker's potential for real-time multimedia applications.

cs.SD↗

arXiv2025

PGSTalker: Real-Time Audio-Driven Talking Head Generation via 3D Gaussian Splatting with Pixel-Aware Density Control

Audio-driven talking head generation is crucial for applications in virtual reality, digital avatars, and film production. While NeRF-based methods enable high-fidelity reconstruction, they suffer from low rendering efficiency and suboptimal audio-visual synchronization. This work presents PGSTalker, a real-time audio-driven talking head synthesis framework based on 3D Gaussian Splatting (3DGS). To improve rendering performance, we propose a pixel-aware density control strategy that adaptively allocates point density, enhancing detail in dynamic facial regions while reducing redundancy elsewhere. Additionally, we introduce a lightweight Multimodal Gated Fusion Module to effectively fuse audio and spatial features, thereby improving the accuracy of Gaussian deformation prediction. Extensive experiments on public datasets demonstrate that PGSTalker outperforms existing NeRF- and 3DGS-based approaches in rendering quality, lip-sync precision, and inference speed. Our method exhibits strong generalization capabilities and practical potential for real-world deployment.

cs.SD↗

arXiv2025

Vehicle-group-based Crash Risk Prediction and Interpretation on Highways

Previous studies in predicting crash risks primarily associated the number or likelihood of crashes on a road segment with traffic parameters or geometric characteristics, usually neglecting the impact of vehicles' continuous movement and interactions with nearby vehicles. Recent technology advances, such as Connected and Automated Vehicles (CAVs) and Unmanned Aerial Vehicles (UAVs) are able to collect high-resolution trajectory data, which enables trajectory-based risk analysis. This study investigates a new vehicle group (VG) based risk analysis method and explores risk evolution mechanisms considering VG features. An impact-based vehicle grouping method is proposed to cluster vehicles into VGs by evaluating their responses to the erratic behaviors of nearby vehicles. The risk of a VG is aggregated based on the risk between each vehicle pair in the VG, measured by inverse Time-to-Collision (iTTC). A Logistic Regression and a Graph Neural Network (GNN) are then employed to predict VG risks using aggregated and disaggregated VG information. Both methods achieve excellent performance with AUC values exceeding 0.93. For the GNN model, GNNExplainer with feature perturbation is applied to identify critical individual vehicle features and their directional impact on VG risks. Overall, this research contributes a new perspective for identifying, predicting, and interpreting traffic risks.

cs.LG↗