Query：
学者姓名：罗欢
Refining：
Year
Type
Indexed by
Source
Complex
Co-
Language
Clean All
Abstract ：
三维点云数据在自动驾驶、机器人和高精地图等领域得到了广泛应用。目前，基于深度学习的三维点云数据处理主要基于有监督学习，其算法性能依赖于大规模高质量的标注数据集。此外，仅在单一设备与场景中训练的三维点云数据处理模型难以应用于不同设备与环境，泛化性能有限。因此，如何减少三维点云标注数据集的需求以及提高三维点云处理模型的适应性是当前三维点云数据处理面临的重要难题。作为迁移学习的一个重要分支，域自适应学习旨在不同域间特征分布存在差异的情况下提高模型的适应性，可为解决上述难题提供重要思路。为便于对点云域自适应学习领域进行更深入有效的探索，本文主要从对抗学习、跨模态学习、伪标签学习、数据对齐及其他方法 5个方面对近年来的三维点云域自适应学习方法进行了系统梳理与分类归纳，并分析总结每类点云域自适应学习方法所具备的优势及面临的问题。最后，对三维点云域自适应学习研究领域的未来发展进行了展望。
Keyword ：
三维点云 三维点云 伪标签学习 伪标签学习 域自适应学习 域自适应学习 对抗学习 对抗学习 数据对齐 数据对齐 跨模态学习 跨模态学习 遥感 遥感
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 范文辉 , 林茜 , 罗欢 et al. 面向三维点云的域自适应学习 [J]. | 遥感学报 , 2024 , 28 (04) : 825-842 . |
| MLA | 范文辉 et al. "面向三维点云的域自适应学习" . | 遥感学报 28 . 04 (2024) : 825-842 . |
| APA | 范文辉 , 林茜 , 罗欢 , 郭文忠 , 汪汉云 , 戴晨光 . 面向三维点云的域自适应学习 . | 遥感学报 , 2024 , 28 (04) , 825-842 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Routing optimization, as a significant part of Traffic Engineering (TE), plays an important role in balancing network traffic and improving quality of service. With the application of Machine Learning (ML) in various fields, many neural network-based routing optimization solutions have been proposed. However, most existing ML-based methods need to retrain the model when confronted with a network unseen during training, which incurs significant time overhead and response delay. To improve the generalization ability of the routing model, in this paper, we innovatively propose a routing optimization method GROM which combines Deep Reinforcement Learning (DRL) and Graph Neural Networks (GNN), to directly generate routing policies under different and unseen network topologies without retraining. Specifically, for handling different network topologies, we transform the traffic-splitting ratio into element -level output of GNN model. To make the DRL agent easier to converge and well generalize to unseen topologies, we discretize the huge continuous trafficsplitting action space. Extensive simulation results on five real-world network topologies demonstrate that GROM can rapidly generate routing policies under different network topologies and has superior generalization ability.
Keyword ：
Graph neural networks Graph neural networks Reinforcement learning Reinforcement learning Software-defined networks Software-defined networks Traffic engineering Traffic engineering
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Ding, Mingjie , Guo, Yingya , Huang, Zebo et al. GROM: A generalized routing optimization method with graph neural network and deep reinforcement learning [J]. | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS , 2024 , 229 . |
| MLA | Ding, Mingjie et al. "GROM: A generalized routing optimization method with graph neural network and deep reinforcement learning" . | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS 229 (2024) . |
| APA | Ding, Mingjie , Guo, Yingya , Huang, Zebo , Lin, Bin , Luo, Huan . GROM: A generalized routing optimization method with graph neural network and deep reinforcement learning . | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS , 2024 , 229 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Three-dimensional (3D) point cloud data have been widely used in many fields, such as autonomous driving, robotics, and high-precision mapping. At present, the state-of-the-art deep learning-based methods for 3D point cloud processing are mainly supervised learning methods. The performance of these methods depends heavily on large-scale, high-quality annotated datasets. However, annotating a large-scale, high-quality, category-diverse, and scenario-rich dataset is time-consuming and labor-intensive. In particular, obtaining sufficiently large numbers of samples for model optimization is also quite difficult in some special cases. In addition, 3D point cloud processing models trained on a single device in a special environment are difficult to generalize to different devices and environments. Their generalizability to various devices and environments is limited. Thus, how to reduce dependencies on high-quality annotated 3D point cloud datasets and how to improve the generalizability of current point cloud processing models are important research topics. In recent years, various kinds of impressive and elaborate technologies, such as meta-learning, few-shot learning, transfer learning, self-supervised learning, semisupervised learning, and weakly supervised learning, have been proposed to solve this problem. As an important research branch of transfer learning, domain adaptive learning aims to eliminate differences in feature distributions across domains and promote the generalization ability of deep learning models, thereby providing a novel solution to address this problem effectively. The academic community has conducted preliminary research on domain adaptive learning for point cloud processing. However, the domain adaptive learning field for point clouds still requires in-depth and effective exploration. Consequently, this study systematically summarizes and classifies recent 3D point cloud domain adaptive learning methods into five categories: adversarial learning, cross-modal learning, pseudo-label learning, data alignment, and other kinds of methods. First, we present the mathematical definition of the domain adaptive learning task and depict the chronological overview of the development of different domain adaptive learning methods to provide readers with a clear understanding. Second, we present the general solution for each category of domain adaptive learning methods and summarize the advantages and disadvantages of the current methods for each category. Third, we compare the performance of current methods on three-point cloud processing tasks, including 3D shape classification, 3D object detection, and 3D semantic segmentation. For each task, we also summarize the commonly used datasets and evaluation metrics for an intuitional comparison. Finally, we conclude the advantages and disadvantages of these five categories of methods and discuss future research directions about the 3D point cloud domain adaptive learning. © 2024 Science Press. All rights reserved.
Keyword ：
3D modeling 3D modeling Deep learning Deep learning Large datasets Large datasets Learning systems Learning systems Remote sensing Remote sensing Supervised learning Supervised learning Transfer learning Transfer learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Fan, Wenhui , Lin, Xi , Luo, Huan et al. Domain adaptation learning for 3D point clouds：A survey [J]. | National Remote Sensing Bulletin , 2024 , 28 (4) : 825-842 . |
| MLA | Fan, Wenhui et al. "Domain adaptation learning for 3D point clouds：A survey" . | National Remote Sensing Bulletin 28 . 4 (2024) : 825-842 . |
| APA | Fan, Wenhui , Lin, Xi , Luo, Huan , Guo, Wenzhong , Wang, Hanyun , Dai, Chenguang . Domain adaptation learning for 3D point clouds：A survey . | National Remote Sensing Bulletin , 2024 , 28 (4) , 825-842 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Hybrid Software Defined Networks (Hybrid SDNs), with a partial upgrade of legacy routers to SDN switches in traditional distributed networks, currently stand as a prevailing network architecture. Traffic Engineering (TE) in hybrid SDN requires the efficient and timely acquisition of a routing policy to adapt to dynamically changing traffic demands, which has recently become a hot topic. Ignoring the hidden relations of consecutive states and the compact representation of the network environment, previous Deep Reinforcement Learning (DRL)-based studies suffer from the convergence problem by only establishing the direct relationship between individual Traffic Matrix (TM) and routing policy. Therefore, to enhance TE performance in hybrid SDNs under dynamically changing traffic demands, we propose to integrate the Transformer model with DRL to establish the relationship between consecutive states and routing policies. The temporal characteristic among consecutive states can effectively assist DRL in solving the convergence problem. To obtain a compact and accurate description of the network environment, we propose to jointly consider TM, routing action, and reward in designing the state of the network environment. To better capture the temporal relations among consecutive states of the network environment, we design a multi-feature embedding module and achieve positional encodings in the Transformer model. The extensive experiments demonstrate that once the convergence problem is solved, the proposed Transformer-based DRL method can efficiently generate routing policies that adapt well to dynamic network traffic.
Keyword ：
Deep Reinforcement Learning Deep Reinforcement Learning Dynamic traffic Dynamic traffic Hybrid SDN Hybrid SDN Traffic Engineering Traffic Engineering Transformer Transformer
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Lin, Bin , Guo, Yingya , Luo, Huan et al. TITE: A transformer-based deep reinforcement learning approach for traffic engineering in hybrid SDN with dynamic traffic [J]. | FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE , 2024 , 161 : 95-105 . |
| MLA | Lin, Bin et al. "TITE: A transformer-based deep reinforcement learning approach for traffic engineering in hybrid SDN with dynamic traffic" . | FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 161 (2024) : 95-105 . |
| APA | Lin, Bin , Guo, Yingya , Luo, Huan , Ding, Mingjie . TITE: A transformer-based deep reinforcement learning approach for traffic engineering in hybrid SDN with dynamic traffic . | FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE , 2024 , 161 , 95-105 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Hybrid Software Defined Networks (Hybrid SDNs), which combines the robustness of distributed network and the flexibility of centralized network, is now a prevailing network architecture. Previous hybrid SDN Traffic Engineering (TE) solutions search an optimal link weight setting or compute the splitting ratios of traffic leveraging heuristic algorithms. However, these methods cannot react timely to the fluctuating traffic demands in dynamic environments and suffer a hefty performance degradation when traffic demands change or network failures happen, especially when network scale is large. To cope with this, we propose a Multi- Agent reinforcement learning based TE method MATE that timely determines the route selection for network flows in dynamic hybrid SDNs. Through dividing the large-scale routing optimization problem into small-scale problem, MATE can better learn the mapping between the traffic demands and routing policy, and efficiently make online routing inference with dynamic traffic demands. To collaborate multiple agents and speed up the convergence in the training process, we innovatively design the actor network and introduce previous actions of all agents in the training of each agent. Extensive experiments conducted on different network topologies demonstrate our proposed method MATE has superior TE performance with dynamic traffic demands and is robust to network failures.
Keyword ：
Dynamic environment Dynamic environment Hybrid Software Defined Networks Hybrid Software Defined Networks Multi-agent reinforcement learning Multi-agent reinforcement learning Traffic Engineering Traffic Engineering
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Guo, Yingya , Ding, Mingjie , Zhou, Weihong et al. MATE: A multi-agent reinforcement learning approach for Traffic Engineering in Hybrid Software Defined Networks [J]. | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS , 2024 , 231 . |
| MLA | Guo, Yingya et al. "MATE: A multi-agent reinforcement learning approach for Traffic Engineering in Hybrid Software Defined Networks" . | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS 231 (2024) . |
| APA | Guo, Yingya , Ding, Mingjie , Zhou, Weihong , Lin, Bin , Chen, Cen , Luo, Huan . MATE: A multi-agent reinforcement learning approach for Traffic Engineering in Hybrid Software Defined Networks . | JOURNAL OF NETWORK AND COMPUTER APPLICATIONS , 2024 , 231 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Since facial forgery techniques have made remarkable progress, the area of forgery detection attracts a significant amount of attention due to security concerns. Existing methods attempt to utilize convolutional neural networks (CNNs) to mine discriminative clues for forgery detection. However, most of these coarse-grained and vanilla methods struggle to extract subtle and multiscale clues in forgery detection. To address such problems, we propose a well-designed deep learning framework, named SCA-Net, to exploit subtle, multiscale and multiview clues. Specifically, our framework consists of a skipped channel attention module (SCM), a constrained difference module (CDM) and an adaptive attention module (AAM). First, the skipped channel attention module is used as the backbone to extract sufficient different information, including low-level and high-level features. Then, the constrained difference module captures manipulation clues from the input image based on constrained characteristics. Finally, the adaptive attention module captures multiscale features represented by facial forgery. Moreover, we introduce a combined loss to address the learning difficulty of our framework. The experimental results demonstrate that the proposed model has great detection performance compared with other face forgery detection methods in most cases. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Su, L. , Wu, B. , Dai, C. et al. Learning to Detect Deepfakes via Adaptive Attention and Constrained Difference [未知]. |
| MLA | Su, L. et al. "Learning to Detect Deepfakes via Adaptive Attention and Constrained Difference" [未知]. |
| APA | Su, L. , Wu, B. , Dai, C. , Luo, H. , Chen, J. . Learning to Detect Deepfakes via Adaptive Attention and Constrained Difference [未知]. |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The use of image-splicing technologies had detrimental effects on the security of multimedia information. Hence, it is necessary to develop effective methods for detecting and locating such tampering. Previous studies have mainly focused on the supervisory role of the mask on the model. The mask edges contain rich complementary signals, which help to fully understand the image and are usually ignored. In this paper, we propose a new network named EAU-Net to detect and locat the splicing regions in the image. The proposed network consists of two parts: Edge-guided SegFormer and Sparse-connected U-Net (SCU). Firstly, the feature extraction module captures local detailed cues and global environment information, which are used to deduce the initial location of the affected regions by SegFormer. Secondly, a Sobel-based edge-guided module (EGM) is proposed to guide the network to explore the complementary relationship between splicing regions and their boundaries. Thirdly, in order to achieve more precise positioning results, SCU is used as postprocessing for removing false alarm pixels outside the focusing regions. In addition, we propose an adaptive loss weight adjustment algorithm to supervise the network training, through which the weights of the mask and the mask edge can be automatically adjusted. Extensive experimental results show that the proposed method outperforms the state-of-the-art splicing detection and localization methods in terms of detection accuracy and robustness.
Keyword ：
Image manipulation Localization Image manipulation Localization Image splicing forgery detection Image splicing forgery detection Splicing detection Splicing detection
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wan, Lin , Su, Lichao , Luo, Huan et al. Combining Edge-Guided Attention and Sparse-Connected U-Net for Detection of Image Splicing [J]. | ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT II , 2023 , 14255 : 167-179 . |
| MLA | Wan, Lin et al. "Combining Edge-Guided Attention and Sparse-Connected U-Net for Detection of Image Splicing" . | ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT II 14255 (2023) : 167-179 . |
| APA | Wan, Lin , Su, Lichao , Luo, Huan , Li, Xiaoyan . Combining Edge-Guided Attention and Sparse-Connected U-Net for Detection of Image Splicing . | ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT II , 2023 , 14255 , 167-179 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Point clouds semantic labeling is an important task in 3D computer vision. Current major researches focus on fully supervised learning. However, point-by-point manual annotations are expensive and time-consuming. To this end, we propose a general point clouds deep active learning framework to ease the annotation burden for researchers. In this work, we propose a conditional random field (CRF) based pseudo labels generation to provide more supervised information for deep neural network (DNN) and employ Golden Loss Correction (GLC) to correct pseudo labeled data training loss. Finally, we propose the Modified Margin acquisition function which can select the most valuable points for labeling. We demonstrate the improvements provided by our proposed method on the S3DIS benchmark.
Keyword ：
Deep active learning Deep active learning Point clouds Point clouds Semantic labeling Semantic labeling
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Hongjing , Luo, Huan , Guo, Wenzhong . A DEEP ACTIVE LEARNING FRAMEWORK FOR SEMANTIC LABELING OF POINT CLOUDS [J]. | IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM , 2023 : 966-969 . |
| MLA | Wu, Hongjing et al. "A DEEP ACTIVE LEARNING FRAMEWORK FOR SEMANTIC LABELING OF POINT CLOUDS" . | IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (2023) : 966-969 . |
| APA | Wu, Hongjing , Luo, Huan , Guo, Wenzhong . A DEEP ACTIVE LEARNING FRAMEWORK FOR SEMANTIC LABELING OF POINT CLOUDS . | IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM , 2023 , 966-969 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Currently, mobile laser scanning (MLS) systems can conveniently and rapidly measure the backscattered laser beam properties of the object surfaces in large-scale roadway scenes. Such properties is digitalized as the in-tensity value stored in the acquired point cloud data, and the intensity as an important information source has been widely used in a variety of applications, including road marking inventory, manhole cover detection, and pavement inspection. However, the collected intensity is often deviated from the object reflectance due to two main factors, i.e. different scanning distances and worn-out surfaces. Therefore, in this paper, we present a new intensity-enhanced method to gradually and efficiently achieve the intensity enhancement in the MLS point clouds. Concretely, to eliminate the intensity inconsistency caused by different scanning distances, the direct relationship between scanning distance and intensity value is modeled to correct the inconsistent intensity. To handle the low contrast between 3D points with different intensities, we proposed to introduce and adapt the dark channel prior for adaptively transforming the intensity information in point cloud scenes. To remove the isolated intensity noises, multiple filters are integrated to achieve the denoising in the regions with different point densities. The evaluations of our proposed method are conducted on four MLS datasets, which are acquired at different road scenarios with different MLS systems. Extensive experiments and discussions demonstrate that the proposed method can exhibit the remarkable performance on enhancing the intensities in MLS point clouds.
Keyword ：
Dark Channel Prior Dark Channel Prior Intensity Enhancement Intensity Enhancement Mobile Laser Scanning Mobile Laser Scanning Point Cloud Point Cloud Point Cloud Denoising Point Cloud Denoising
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Fang, Lina , Chen, Hao , Luo, Huan et al. An intensity-enhanced method for handling mobile laser scanning point clouds [J]. | INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION , 2022 , 107 . |
| MLA | Fang, Lina et al. "An intensity-enhanced method for handling mobile laser scanning point clouds" . | INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 107 (2022) . |
| APA | Fang, Lina , Chen, Hao , Luo, Huan , Guo, Yingya , Li, Jonathon . An intensity-enhanced method for handling mobile laser scanning point clouds . | INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION , 2022 , 107 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Feature matching is a fundamental problem in feature-based remote sensing image registration. Due to the ground relief variations and imaging viewpoint changes, remote sensing images often involve local distortions, leading to difficulties in high-accuracy image registration. To address this issue, in this article, we propose a robust feature matching method called First Neighbor Relation Guided (FNRG) for remote sensing image registration via guided hyperplane fitting. The key idea of FNRG is to exploit the first neighbor relation of feature points between two images for seeking consistent seeds in a parameter-free manner. To boost more consistent matches based on the consistent seeds, we formulate the feature matching problem into an affine hyperplane fitting problem by imposing the motion consistency, and then we design a hyperplane updating strategy to refine the fitting model. We also introduce a locality preserving structure-based cost function to promote the matching performance of the hyperplane updating strategy. Our method can mine consistent matches from thousands of putative ones within only a few milliseconds, and it also can handle the data with a large-scale change, rotation, or severe nonrigid deformation. Extensive experiments on the remote sensing image data sets with different types of image transformations show that the proposed method achieves significant superiority over several state-of-the-art methods.
Keyword ：
Data models Data models Distortion Distortion Feature matching Feature matching first neighbor relation first neighbor relation guided feature matching guided feature matching Image registration Image registration Imaging Imaging remote sensing remote sensing Remote sensing Remote sensing Sensors Sensors Strain Strain
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xiao, Guobao , Luo, Huan , Zeng, Kun et al. Robust Feature Matching for Remote Sensing Image Registration via Guided Hyperplane Fitting [J]. | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING , 2022 , 60 . |
| MLA | Xiao, Guobao et al. "Robust Feature Matching for Remote Sensing Image Registration via Guided Hyperplane Fitting" . | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 60 (2022) . |
| APA | Xiao, Guobao , Luo, Huan , Zeng, Kun , Wei, Leyi , Ma, Jiayi . Robust Feature Matching for Remote Sensing Image Registration via Guided Hyperplane Fitting . | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING , 2022 , 60 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
