A5000791743- Robotics and Sensor-Based Localization
- Augmented Reality Applications
- IoT and Edge/Fog Computing
- Visual Attention and Saliency Detection
- Advanced Neural Network Applications
- UAV Applications and Optimization
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Advanced Data Compression Techniques
- Network Time Synchronization Technologies
- Image Retrieval and Classification Techniques
- Complex Network Analysis Techniques
- Recommender Systems and Techniques
- Smart Grid Security and Resilience
- Advanced Graph Neural Networks
- Robotic Path Planning Algorithms
- Opinion Dynamics and Social Influence
- Virtual Reality Applications and Impacts
- Software-Defined Networks and 5G
Tsinghua University
2021-2024
Software (Spain)
2024
Chongqing University
2022
CNN based real-time object detection can facilitate various AI applications that need to understand the surroundings via camera, such as autonomous package delivery robots, augmented reality, and intelligent drone applications. Currently, due high computation cost of CNN, accurate is only possible when mobile devices upload video frames powerful edge servers through high-speed wireless networks like WiFi. However, for many open-air applications, network conditions (such cellular networks)...
Real-time mobile video analysis like object detection and tracking is key to various household applications such as AR, cognitive assistance smart home. Such rely on heavy DNN models, which are not suitable for devices due resource limitation. The long latency of cloud offloading unacceptable the real-time requirements, direct edge relies powerful servers, impractical scenarios. To solve this challenge, we take advantage computing that low-cost or already exist in our lives, propose Trine, a...
Most existing multi-user Augmented Reality (AR) systems only support multiple co-located users to view a common set of virtual objects but lack the ability enable each user directly interact with other appearing in his/her view. Such AR should be able detect human keypoints and estimate device poses (for identifying different users) meantime. However, due stringent low latency requirements intensive computation preceding two capabilities, previous research enables either capabilities for...
Most existing multi-user Augmented Reality (AR) systems only support multiple co-located users to view a common set of virtual objects but lack the ability enable each user directly interact with other appearing in his/her view. Such AR should be able detect human keypoints and share device poses (for identifying different users) meanwhile. However, due stringent low latency requirements intensive computation above two capabilities, previous research enables either capabilities for mobile...
Drones have witnessed extensive popularity among diverse smart applications, and visual SLAM technology is commonly used to estimate 6-DoF pose for the drone flight control system. However, traditional image-based cannot ensure safety of drones, especially in challenging environments such as high-speed high dynamic range scenarios. Event camera, a new vision sensor, holds potential enable drones overcome above scenarios if fused into SLAM. Unfortunately, computational demands event-image...
Increasing the size of embedding layers has shown to be effective in improving performance recommendation models, yet gradually causing their sizes exceed terabytes industrial recommender systems, and hence increase computing storage costs. To save resources while maintaining model performances, we propose SHARK, compression practice have summarized system scenarios. SHARK consists two main components. First, use novel first-order component Taylor expansion as importance scores prune number...
Mobile multiplayer augmented reality(AR) emerges in various applications including games, education, training, etc. Edge computing technique enables real-time environmental perception ability(e.g. object detection and segmentation) of devices by offloading complex computation to the nearby edge server. However, with more players involved video streams, bandwidth competition intensifies lengthens transmission latency, which severely impairs accuracy AR applications. We realize staggering...