A5053026785- Access Control and Trust
- Privacy-Preserving Technologies in Data
- Peer-to-Peer Network Technologies
- Advanced Neural Network Applications
- Cryptography and Data Security
- Industrial Vision Systems and Defect Detection
- Caching and Content Delivery
- Network Security and Intrusion Detection
- Software System Performance and Reliability
- 3D Surveying and Cultural Heritage
- Advanced Graph Neural Networks
- Human Pose and Action Recognition
- Robotics and Sensor-Based Localization
- Data Management and Algorithms
- Wikis in Education and Collaboration
- Open Source Software Innovations
- Online Learning and Analytics
- Software-Defined Networks and 5G
- Knowledge Management and Sharing
- Distributed systems and fault tolerance
- Brain Tumor Detection and Classification
- Advanced Computing and Algorithms
- Cloud Data Security Solutions
- Remote Sensing and LiDAR Applications
- Mobile Crowdsensing and Crowdsourcing
Tongji University
2008-2024
Beijing University of Posts and Telecommunications
2007-2012
RGB-D semantic segmentation can be advanced with convolutional neural networks due to the availability of Depth data. Although objects cannot easily discriminated by just 2D appearance, local pixel difference and geometric patterns in Depth, they well separated some cases. Considering fixed grid kernel structure, CNNs are limited lack ability capture detailed, fine-grained information thus achieve accurate pixel-level segmentation. To solve this problem CNN we propose a Pixel Difference...
Open source has revolutionized how software development is carried out, with a growing number of individuals and organizations contributing to open projects. As the importance continues grow, companies also expect grow thriving sustainable communities continued contributions better collaborations. In this study, we applied contribution leaderboard seven projects initiated by Alibaba. We conducted case study investigate perceptions facts regarding motivate collaboration through gamification....
Abstract Federated learning (FL) is a technology that allows multiple devices to collaboratively train global model without sharing original data, which hot topic in distributed intelligent systems. Combined with satellite network, FL can overcome the geographical limitation and achieve broader applications. However, it also faces issues such as straggler effect, unreliable network environments non-independent identically (Non-IID) samples. To address these problems, we propose an...
Combining RGB images and the corresponding depth maps in semantic segmentation proves effectiveness past few years. Existing RGB-D modal fusion methods either lack non-linear feature ability or treat both equally, regardless of intrinsic distribution gap information loss. Here we find that are suitable to provide fine-grained patterns objects due their local continuity, while effectively a global view. Based on this, propose pixel differential convolution attention (DCA) module consider...
The great challenges of Internet service fault management are uncertainty and noise. To address these challenges, we model the scenario through a multi-layer model, propose an approach using active probing to detect diagnose faults. This uses bipartite Bayesian network as dependency binary symmetric channel noise is composed two phases: detection diagnosis. In first phase, greedy approximation probe selection algorithm (GAPSA), which selects minimal set probes while remaining high...
A novel reputation-based trust model for P2P networks is presented in this paper. The aims to measure the credibility of peerspsila referrals, and thus filter noise responses obtain more accurate values. Furthermore reputation value concretely quantified by a recommenderpsilas credibility. Simulation results show that our advanced successful transaction rate than other model.
Managing trust is a key issue for wide acceptance of P2P computing, particularly in critical areas such as e-commerce. Reputation based management has been identified the literature viable solution to problem. However, mechanism faces challenges subjectively, experiential weighting referrals when aggregating recommendation information. Furthermore, not considering some malicious attacks building relationship between peers existing schemes make model very vulnerable. This paper presents...
Neural implicit representations have recently demonstrated considerable potential in the field of visual simultaneous localization and mapping (SLAM). This is due to their inherent advantages, including low storage overhead representation continuity. However, these methods necessitate size scene as input, which impractical for unknown scenes. Consequently, we propose NeB-SLAM, a neural block-based scalable RGB-D SLAM Specifically, first divide-and-conquer strategy that represents entire set...
3D Gaussian Splatting has emerged as a promising technique for high-quality rendering, leading to increasing interest in integrating 3DGS into realism SLAM systems. However, existing methods face challenges such primitives redundancy, forgetting problem during continuous optimization, and difficulty initializing monocular case due lack of depth information. In order achieve efficient photorealistic mapping, we propose RP-SLAM, splatting-based vision method RGB-D cameras. RP-SLAM decouples...
With the growth of web technology, semantic offers a promising framework for online knowledge collaboration. However, trust issues can undermine users' willingness to collaborate, reduce frequency interaction and collaboration efficiency. This paper introduces super node-based management model designed enhance networks by linking nodes through relationships. The exploits synergistic incentives similar interest behaviours achieve steady construction We propose similarity filtering algorithm...
Abstract In this paper, we develop a delay‐centric parallel multi‐path routing protocol for multi‐hop cognitive radio ad hoc networks. First, analyze the end‐to‐end delay of based on queueing theory and present new dynamic traffic assignment scheme with objective minimizing delay, considering both spectrum availability link data rate. The problem is formulated as convex solved by gradient‐based search method to obtain optimal assignments. Furthermore, heuristic decentralized presented. Then,...