A5007137713- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Autonomous Vehicle Technology and Safety
- Video Surveillance and Tracking Methods
- Speech and Audio Processing
- Biometric Identification and Security
- IoT and Edge/Fog Computing
- User Authentication and Security Systems
- Advanced SAR Imaging Techniques
- Age of Information Optimization
- Mobile Ad Hoc Networks
- Music and Audio Processing
- Remote Sensing and LiDAR Applications
- Face recognition and analysis
- Speech Recognition and Synthesis
- Wireless Body Area Networks
- Automated Road and Building Extraction
- Vehicular Ad Hoc Networks (VANETs)
Chinese University of Hong Kong
2021-2024
Infrastructure-assisted autonomous driving is an emerging paradigm that expects to significantly improve the safety of vehicles. The key enabling technology for this vision fuse LiDAR results from roadside infrastructure and vehicle vehicle's perception in real time. In work, we propose VIPS, a novel lightweight system can achieve decimeter-level real-time (up 100 ms) fusion between vehicles infrastructure. idea VIPS exploit highly efficient matching graph structures encode objects' lean...
A wide range of advanced deep learning algorithms have recently been proposed for image classification and object detection. However, the effectiveness these methods can be significantly restricted in many real-world scenarios where visibility or illumination is poor. Compared to RGB cameras, millimeter-wave (mmWave) radars are immune above environmental variability assist cameras under adverse conditions. To this end, we propose milliEye, a lightweight mmWave radar camera fusion system...
Head-mounted wearables are rapidly growing in popularity. However, a gap exists providing robust voice-related applications like conversation or command control complex environments, such as competing speakers and strong noises. The compact design of HMWs introduces non-trivial challenges to existing speech enhancement systems that use microphone recording only. In this paper, we handle problem by using bone vibration conducted through the head skull. principle is accelerometer widely...
Recently, smart roadside infrastructure (SRI) has demonstrated the potential of achieving fully autonomous driving systems. To explore infrastructure-assisted driving, this paper presents design and deployment Soar, first end-to-end SRI system specifically designed to support Soar consists both software hardware components carefully overcome various physical challenges. can leverage existing operational like street lampposts for a lower barrier adoption. adopts new communication architecture...
HD map is a key enabling technology towards fully autonomous driving. We propose VI-Map, the first system that leverages roadside infrastructure to enhance real-time mapping for The core concept of VI-Map exploit unique cumulative observations made by build and maintain an accurate current map. This then fused with on-vehicle maps in real time, resulting more comprehensive up-to-date By extracting concise bird-eye-view features from utilizing vectorized representations, incurs low compute...
Recent years have witnessed the emergence of a new class cooperative edge systems in which large number nodes can collaborate through local peer-to-peer connectivity. In this paper, we propose CoEdge, novel system that support concurrent data/compute-intensive deep learning (DL) models for distributed real-time applications such as city-scale traffic monitoring and autonomous driving. First, CoEdge includes hierarchical DL task scheduling framework dispatches tasks to based on their...
Infrastructure-assisted autonomous driving is an emerging paradigm that expects to significantly improve the safety of vehicles. The key enabling technology for this vision fuse LiDAR results from roadside infrastructure and vehicle vehicle's perception in real time. In work, we propose VIPS, a novel lightweight system can achieve decimeter-level real-time (up 100ms) fusion between vehicles infrastructure. idea VIPS exploit highly efficient matching graph structures encode objects' lean...
In recent years, the deployment of unmanned vehicle delivery services has increased unprecedentedly, leading to a need for enhanced security due risk leaving high-value packages an unauthorized third party during pickup or delivery. Existing authentication methods such as QR code and one-time password are inadequate, they susceptible attacks provide only one-way authentication. This paper, first time our best knowledge, proposes Wave-for-Safe (W4S) --- novel mutual system that utilizes...
Recently,smart roadside infrastructure (SRI) has demonstrated the potential of achieving fully autonomous driving systems. To explore infrastructure-assisted driving, this paper presents design and deployment Soar, first end-to-end SRI system specifically designed to support Soar consists both software hardware components carefully overcome various physical challenges. can leverage existing operational like street lampposts for a lower barrier adoption. adopts new communication architecture...