Filters








47,549 Hits in 2.4 sec

Multi-scale Dynamic Graph Convolution Network for Point Clouds Classification

ZhengLi Zhai, Xin Zhang, LuYao Yao
2020 IEEE Access  
In this paper, we propose a Multi-scale Dynamic GCN model for point clouds classification, a Farthest Point Sampling method is applied in our model firstly to efficiently cover the entire point set, it  ...  uses different scale k-NN group method to locate on k nearest neighborhood for each central node, Edge Convolution (EdgeConv) operation is used to extract and aggregate local features between neighbor  ...  TRAINING We evaluated our Multi-scale Dynamic GCN model onMod-elNet10 & ModelNet40 dataset classification task.  ... 
doi:10.1109/access.2020.2985279 fatcat:s74mrb4a3fdmfdzqstejglchli

Multi-scale Dynamic Graph Convolutional Network for Hyperspectral Image Classification [article]

Sheng Wan and Chen Gong and Ping Zhong and Bo Du and Lefei Zhang and Jian Yang
2019 arXiv   pre-print
Therefore, our method is termed 'Multi-scale Dynamic Graph Convolutional Network' (MDGCN).  ...  To alleviate this shortcoming, we consider employing the recently proposed Graph Convolutional Network (GCN) for hyperspectral image classification, as it can conduct the convolution on arbitrarily structured  ...  CONCLUSION In this paper, we propose a novel Multi-scale Dynamic Graph Convolutional Network (MDGCN) for hyperspectral image classification.  ... 
arXiv:1905.06133v1 fatcat:wn33smzhrfaopp62ecinqsae3q

Dynamic Multi-scale Convolutional Neural Network for Time Series Classification

Bin Qian, Yong Xiao, Zhenjing Zheng, Mi Zhou, Wanqing Zhuang, Sen Li, Qianli Ma
2020 IEEE Access  
In this paper, we propose dynamic multi-scale convolutional neural network to extract multi-scale feature representations existing in each time series dynamically.  ...  INDEX TERMS Convolutional neural networks, multi-scale temporal features, time series classification.  ...  FIGURE 3 . 3 General architecture of the Dynamic Multi-Scale Convolutional Neural Network (DMS-CNN).  ... 
doi:10.1109/access.2020.3002095 fatcat:wc3s66qke5fwjo6n7rrw4ql5h4

Dynamic Sampling Network for Semantic Segmentation

Bin Fu, Junjun He, Zhengfu Zhang, Yu Qiao
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Moreover, we utilize the multi-scale contextual representations to guide the sampling process.  ...  Therefore, our CGDS can adaptively capture shape and scale information according to not only the input feature map but also the multi-scale semantic context.  ...  It employs multi-scale contextual information to generate sampling points for low level feature map.  ... 
doi:10.1609/aaai.v34i07.6709 fatcat:utu324fjrzbudjw6d7e5wa7xce

Open-Narrow-Synechiae Anterior Chamber Angle Classification in AS-OCT Sequences [article]

Huaying Hao, Huazhu Fu, Yanwu Xu, Jianlong Yang, Fei Li, Xiulan Zhang, Jiang Liu, Yitian Zhao
2020 arXiv   pre-print
To address this, we propose a novel sequence multi-scale aggregation deep network (SMA-Net) for open-narrow-synechiae ACA classification based on an AS-OCT sequence.  ...  In our method, a Multi-Scale Discriminative Aggregation (MSDA) block is utilized to learn the multi-scale representations at slice level, while a ConvLSTM is introduced to study the temporal dynamics of  ...  To be more specific, we introduced a novel block, named the MSDA block, with a view to learning multi-scale discriminative representations over AS-OCT volumes.  ... 
arXiv:2006.05367v1 fatcat:nqpak4yo65bidg24d6hv2x73vi

A Dynamic Multi-Scale Network for EEG Signal Classification

Guokai Zhang, Jihao Luo, Letong Han, Zhuyin Lu, Rong Hua, Jianqing Chen, Wenliang Che
2021 Frontiers in Neuroscience  
To address this challenge, in this paper, we propose a novel dynamic multi-scale network to achieve the EEG signal classification.  ...  ability, we incorporate a dynamic multi-scale (DMS) layer, which allows the network to learn multi-scale features from different receptive fields at a more granular level.  ...  To handle this problem, in this paper, we propose a dynamic multi-scale network for the EEG signal classification.  ... 
doi:10.3389/fnins.2020.578255 pmid:33519352 pmcid:PMC7838674 fatcat:jb7m6hlkirgnzlvszh2pshkpia

Stack Attention-Pruning Aggregates Multiscale Graph Convolution Networks for Hyperspectral Remote Sensing Image Classification

Na Liu, Bin Zhang, Qiuhuan Ma, Qingqing Zhu, Xiaoling Liu
2021 IEEE Access  
Then we adopt the aggregation manner for multiscale graph convolution of pixels nodes in different neighborhood for effective long-range joint interaction modeling.  ...  INDEX TERMS Hyperspectral remote sensing image classification, stack attention-pruning, multiscale graph convolution networks, longdistances joint interaction, multiscale spatial-temporal information,  ...  By contrast, GCN-based methods such as multi scale dynamic graph convolution networks (MDGCN) [26] . C.  ... 
doi:10.1109/access.2021.3061489 fatcat:jqsobopyxnhb7ptvcldvepwk5e

Fully Convolutional Online Tracking [article]

Yutao Cui, Cheng Jiang, Limin Wang, Gangshan Wu
2021 arXiv   pre-print
Based on the online RGM, we devise a simple anchor-free tracker (FCOT), composed of a feature backbone, an up-sampling decoder, a multi-scale classification branch, and a multi-scale regression branch.  ...  To tackle this issue, we present the fully convolutional online tracking framework, coined as FCOT, and focus on enabling online learning for both classification and regression branches by using a target  ...  REG-2S REPRESENTS FOR MULTI-SCALE REGRESSION HEAD, CLS-2S FOR MULTI-SCALE CLASSIFICATION, UP FOR UP-SAMPLE BLOCK AND ONLINE-REG FOR ONLINE REGRESSION MODEL GENERATOR.  ... 
arXiv:2004.07109v5 fatcat:lxwkgvz73vejrnjbgn6ydus4cu

Dynamic Multi-scale Convolution for Dialect Identification [article]

Tianlong Kong, Shouyi Yin, Dawei Zhang, Wang Geng, Xin Wang, Dandan Song, Jinwen Huang, Huiyu Shi, Xiaorui Wang
2021 arXiv   pre-print
To address this issue, we propose a new architecture, named dynamic multi-scale convolution, which consists of dynamic kernel convolution, local multi-scale learning, and global multi-scale pooling.  ...  Local multi-scale learning, which represents multi-scale features at a granular level, is able to increase the range of receptive fields for convolution operation.  ...  Dynamic Multi-scale Convolution for Dialect Identification In this section, we describe the proposed dynamic multi-scale convolution method for dialect identification.  ... 
arXiv:2108.07787v1 fatcat:cfykz4cfnrbbzn4l66cv2cwzx4

Dynamic Temporal Pyramid Network: A Closer Look at Multi-Scale Modeling for Activity Detection [article]

Da Zhang, Xiyang Dai, Yuan-Fang Wang
2019 arXiv   pre-print
input video frames dynamically with varying frame per seconds (FPS) to construct a natural pyramidal input for video of an arbitrary length. (2) We design a two-branch multi-scale temporal feature hierarchy  ...  While spatial multi-scale modeling has been well studied in object detection, how to effectively apply a multi-scale architecture to temporal models for activity detection is still under-explored.  ...  The localization branch uses temporal convolution for better localization while the classification branch uses maximum pooling to record the most prominent features for recognition.  ... 
arXiv:1808.02536v2 fatcat:fyvlzksc5vfmxebch2iy6uvtgu

MS-CapsNet: A Novel Multi-Scale Capsule Network

Canqun Xiang, Lu Zhang, Yi Tang, Wenbin Zou, Chen Xu
2018 IEEE Signal Processing Letters  
However, the original capsule network is not suitable for some classification tasks that the detected object has complex internal representations.  ...  The proposed Multi-Scale Capsule Network consists of two stages. In the first stage the structural and semantic information are obtained by the multi-scale feature extraction.  ...  The multi-scale convolution feature extraction and multi-dimensional capsule coding is employed to learn rich represents.  ... 
doi:10.1109/lsp.2018.2873892 fatcat:y5bq4xwxi5b7loa5etwggvmn44

Multi-scale Hybrid Pooling Convolutional Neural Network Algorithm

Nan ZHAO, Xin WANG, Ying-na LI, Sheng WU
2018 DEStech Transactions on Engineering and Technology Research  
In this paper, we propose a multi-scale hybrid pooling algorithm based on convolutional neural network model, through convolution.  ...  Learning Algorithm Principle and Description The Multi-scale Hybrid Pooling Convolutional Neural Network (MSHP-CNN) model proposed in this chapter is an improved model based on the CNN model.  ...  In the pooling layer, unlike the traditional network structure, the multi-scale hybrid pooled convolutional neural network model adopts a mechanism of dynamic adjustment based on image information distribution  ... 
doi:10.12783/dtetr/ecar2018/26369 fatcat:zqjzqpmrbjagveug5nxfgfddnu

Densely Connected CNN with Multi-scale Feature Attention for Text Classification

Shiyao Wang, Minlie Huang, Zhidong Deng
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
Furthermore, a multi-scale feature attention is developed to adaptively select multi-scale features for classification.  ...  In this paper, we present a densely connected CNN with multi-scale feature attention for text classification.  ...  By multi-scale feature attention, the model can adaptively select task-friendly yet effective features from many multi-scale features for text classification.  ... 
doi:10.24963/ijcai.2018/621 dblp:conf/ijcai/WangHD18 fatcat:zx6aa5sx25d2pablints6c7gce

StructToken : Rethinking Semantic Segmentation with Structural Prior [article]

Fangjian Lin, Zhanhao Liang, Junjun He, Miao Zheng, Shengwei Tian, Kai Chen
2022 arXiv   pre-print
In this paper, we present structure token (StructToken), a new paradigm for semantic segmentation.  ...  We hope that structure token could serve as an alternative for semantic segmentation and inspire future research.  ...  model, and +1.33 % mIoU (40.23 vs. 38 .90) for multi-scale inference.  ... 
arXiv:2203.12612v3 fatcat:kof2zssz6bfdfj2jjit6lm434i

Multi-hop Graph Convolutional Network with High-order Chebyshev Approximation for Text Reasoning [article]

Shuoran Jiang, Qingcai Chen, Xin Liu, Baotian Hu, Lisai Zhang
2021 arXiv   pre-print
In this study, we define the spectral graph convolutional network with the high-order dynamic Chebyshev approximation (HDGCN), which augments the multi-hop graph reasoning by fusing messages aggregated  ...  from direct and long-term dependencies into one convolutional layer.  ...  Conclusions This study proposes a multi-hop graph convolutional network on high-order dynamic Chebyshev approximation (HDGCN) for text reasoning.  ... 
arXiv:2106.05221v1 fatcat:5cw563x46fbhxmeoe35xxzz6qm
« Previous Showing results 1 — 15 out of 47,549 results