A5043353686- Remote Sensing and LiDAR Applications
- 3D Surveying and Cultural Heritage
- 3D Shape Modeling and Analysis
- Domain Adaptation and Few-Shot Learning
- Cryptography and Data Security
- COVID-19 diagnosis using AI
- Multimodal Machine Learning Applications
- Cloud Data Security Solutions
- Privacy-Preserving Technologies in Data
- Blockchain Technology Applications and Security
- Face and Expression Recognition
- Advanced Clustering Algorithms Research
- Stochastic Gradient Optimization Techniques
- Fuzzy Logic and Control Systems
Jianghan University
2022-2023
Federated learning in mobile edge computing allows a larger number of devices to jointly train an accurate machine model without collecting local data from nodes. However, there are two major challenges using federated for computing. The first is that networks only tolerate limited communication overhead, is, overhead between nodes, servers, and cloud servers cannot be excessive. Unfortunately, clients send large update do not meet realistic requirements. second resource-constrained nodes...
Generalized zero-shot learning (GZSL) aims to solve the category recognition tasks for unseen categories under setting that training samples only contain seen classes while are not available. This research is vital as there always existing new and large amounts of unlabeled data in realistic scenarios. Previous work GZSL usually maps visual information visible semantic description invisible into identical embedding space bridge gap between disjointed classes, ignoring intrinsic features...
PointNet is a deep neural network that directly takes 3D point cloud data as inputs. Due to its strong stability and computational efficiency, has become one of the most popular classification methods in real applications. Recently, transformer techniques have achieved great successes classifying objects with image inputs, which inspires us transplant transformers into object In this paper, we propose method based on transformers. Firstly, an offset-attention module added after spatial...
Self-organizing maps (SOM) have become a commonly-used cluster analysis technique in data mining. However, SOM are not able to process incomplete data. To build more capability of mining for SOM, this study proposes an SOM-based fuzzy map model with sets. Using model, translated into data, and used generate observations. These observations, along observations without missing values, then train the maps. Compared standard approach, generated by proposed method can provide information...
With the widespread adoption of cloud storage, ensuring integrity outsourced data has become increasingly important. Various storage auditing protocols based on public key cryptography have been proposed. However, all them require complex cryptographic operations and incur significant communication costs. To address issues overhead for tags, high computational complexity algorithms, limited efficiency dynamic algorithms in signature algorithm-based outsourcing verification protocols, we...
Generalized zero-shot learning (GZSL) aims to solve the category recognition tasks for unseen categories under setting that training samples only contain seen classes while are not available. This research is vital as there always existing new and large amounts of unlabeled data in realistic scenarios. Previous work GZSL usually maps visual information visible semantic description invisible into identical embedding space bridge gap between disjointed classes, ignoring intrinsic features...