检索结果-内蒙古大学图书馆

Proceedings of the 37th International Conference on Neural Information processing Systems

作者： Shaolei Zhang Yang Feng Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences (ICT/CAS) and University of Chinese Academy of Sciences

Simultaneous sequence generation is a pivotal task for real-time scenarios, such as streaming speech recognition, simultaneous machine translation and simultaneous speech translation, where the target sequence is generated while receiving the source sequence. The crux of achieving high-quality generation with low latency lies in identifying the optimal moments for generating, accomplished by learning a mapping between the source and target sequences. However, existing methods often rely on task-specific heuristics for different sequence types, limiting the model's capacity to adaptively learn the source-target mapping and hindering the exploration of multi-task learning for various simultaneous tasks. In this paper, we propose a unified segment-to-segment framework (Seg2Seg) for simultaneous sequence generation, which learns the mapping in an adaptive and unified manner. During the process of simultaneous generation, the model alternates between waiting for a source segment and generating a target segment, making the segment serve as the natural bridge between the source and target. To accomplish this, Seg2Seg introduces a latent segment as the pivot between source to target and explores all potential source-target mappings via the proposed expectation training, thereby learning the optimal moments for generating. Experiments on multiple simultaneous generation tasks demonstrate that Seg2Seg achieves state-of-the-art performance and exhibits better generality across various tasks. Code is available at: https://***/ictnlp/Seg2Seg.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Boundary Cue Guidance and Contextual Feature Mining for Glasss Segmentation

Boundary Cue Guidance and Contextual Feature Mining for Glas...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Qiquan Xiao Yuan Zhang Xuanya Li Kai Hu Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Xiangtan China Baidu Inc. Beijing China

Glass is ubiquitous in the real world, and its perception has many applications, including robot navigation and drone tracking. However, due to the transparent property of glass, the interior of a glass area can be any surrounding scene or object, which brings challenges for computer vision. Inspired by the human senses, boundary cues are one of the crucial factors for people to judge the location of glass contours. Hence, we propose a boundary cue guidance and contextual feature mining network (BCNet) to accurately and efficiently segment glass. Specifically, we first design a multi-branch boundary extraction module (MBEM) for learning accurate boundary cues combined with multi-level encoded features. Second, we propose a boundary cue guidance module (BCGM), inject the boundary cues into the representation learning, and provide constraints with object structure semantics to guide feature extraction. Besides, we design a contextual feature mining module (CFMM) to dynamically capture the contextual information of different receptive fields for the detection of different sizes and shapes of the glass. Finally, extensive experiments on two benchmark glass datasets, GDD and GSD. The results demonstrate that our BCNet achieves state-of-the-art segmentation performance against existing methods.

关键词： Representation learning Shape Semantics Neural networks Glass Benchmark testing Signal processing

来源：评论

学校读者我要写书评

暂无评论

Object Detection Algorithm Based on Second-Order Pooling Network and Gaussian Mixture Attention 5

Object Detection Algorithm Based on Second-Order Pooling Net...

引用

5th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2022

作者： Ma, Sugang Li, Ningbo Yang, Xiaobao School of Computer Science and Technology Xi'an University of Posts and Telecommunications Shaanxi Xi'an China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications Shaanxi Xi'an China Xi'an Key Laboratory of Big Data and Intelligent Computing Xi'an University of Posts and Telecommunications Shaanxi Xi'an China

ISBN: (纸本)9781450396899

To improve the feature representation ability of the YOLOX algorithm and obtain better detection performance, an object detection algorithm based on second-order pooling network and gaussian mixture attention is proposed. Firstly, the second-order pooling network is added after the PAFPN, and the higher-order statistical information is obtained by calculating the covariance matrix between different channels, to enhance the non-linear modeling capability. Secondly, the mixture attention based on the gaussian function is added after the SPP to model global contexts in the spatial and channel dimensions respectively, which improves the network performance with almost no extra parameters. The experimental results show that the detection accuracy of the proposed algorithm on the PASCAL VOC dataset reaches 82.6 %, which is 1.6 % higher than the YOLOX algorithm. © 2022 ACM.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Neural Architecture Search for Transformer in Knowledge Tracing

arXiv

引用

arXiv 2023年

作者： Yang, Shangshang Yu, Xiaoshan Tian, Ye Yan, Xueming Ma, Haiping Zhang, Xingyi Anhui University China Guangdong University of Foreign Studies China Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education China

Transformer has achieved excellent performance in the knowledge tracing (KT) task, but they are criticized for the manually selected input features for fusion and the defect of single global context modelling to directly capture students’ forgetting behavior in KT, when the related records are distant from the current record in terms of time. To address the issues, this paper first considers adding convolution operations to the Transformer to enhance its local context modelling ability used for students’ forgetting behavior, then proposes an evolutionary neural architecture search approach to automate the input feature selection and automatically determine where to apply which operation for achieving the balancing of the local/global context modelling. In the search space design, the original global path containing the attention module in Transformer is replaced with the sum of a global path and a local path that could contain different convolutions, and the selection of input features is also considered. To search the best architecture, we employ an effective evolutionary algorithm to explore the search space and also suggest a search space reduction strategy to accelerate the convergence of the algorithm. Experimental results on the two largest and most challenging education datasets demonstrate the effectiveness of the architecture found by the proposed approach. Copyright © 2023, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

SKT-MOT and DyTracker: A Multiobject Tracking Dataset and a Dynamic Tracker for Speed Skating Video

引用

Scientific Programming 2023年第1期2023卷

作者： Wang, Junwu Li, Zongmin Li, Yachuan Yang, Shaobo Wang, Ben Li, Hua Qingdao266580 China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China

Speed skating serves as a significant application domain for multiobject tracking (MOT), presenting unique challenges such as frequent occlusion, highly similar appearances, and motion blur. To address these challenges, this paper constructs an MOT dataset called SKT-MOT for speed skating and analyzes the shortcomings of existing datasets and methods. Accordingly, we propose a dynamic MOT method called DyTracker. The method builds upon the DeepSORT baseline and enhances three key modules. At the global level, we design the track dynamic management (TDM) algorithm. In the motion branch, a novel metric is proposed to evaluate occlusion and Kalman filter dynamic update (KFDU) is implemented. In the appearance branch, we account for the difference in human posture and propose the feature dynamic selection and updating (FDSU) strategy. This makes our DyTracker flexible and efficient to achieve a multiobject tracking accuracy (MOTA) of 93.70% and identification F1 (IDF1) score of 92.39% on SKT-MOT, which is a significant advantage over existing SOTA methods. To validate the generalization of our proposed module, two dynamic update modules are inserted into other methods and validated on the public dataset MOT17, and the accuracy is generally improved by 0.2%-0.6%. © 2023 Junwu Wang et al.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Subnetwork Reliability Analysis About Complete-Transposition Graph Networks

SSRN

引用

SSRN 2023年

作者： Chen, Qun Deng, Qingying Xiang, Kainan Key Laboratory of Intelligent Computing & Information Processing of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

With the size and complexity of a multiprocess computer system grows, the likelihood of having faulty processors in the system increases. How to evaluate the impact of faulty processors on the entire system is what we care about. Reliability evaluation is an important indicator to measure the faulty's effect on the entire system. A typical approach to evaluate the reliability of a system is using the probability that a fault-free subsystem of a specific size is still working in the system when the system has some faulty processors. The higher the probability is the more reliable the system is. In this paper, we will use the probability fault model and the principle of Inclusion-Exclusion to establish the CTn−1 subnetwork reliability of CTn in the case of node failure. An upper bound and lower bound are derived by taking into account the intersection of no more than four subnetworks. In addition, we show that the theoretical results are proved to be in accordance with near to the simulation results, especially when the value of single-node reliability p goes low. © 2023, The Authors. All rights reserved.

关键词： Reliability analysis

来源：评论

学校读者我要写书评

暂无评论

Transwnet: Integrating Transformers into CNNS via Row and Column Attention for Abdominal Multi-Organ Segmentation 48

Transwnet: Integrating Transformers into CNNS via Row and Co...

引用

48th IEEE International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Xie, Yazhen Huang, Yanglin Zhang, Yuan Li, Xuanya Ye, Xiongjun Hu, Kai Xiangtan University Key Laboratory of Intelligent Computing Information Processing of Ministry of Education Xiangtan411105 China Baidu Inc. Beijing100085 China Chinese Academy of Medical Sciences and Peking Union Medical College National Cancer Center Cancer Hospital Beijing100021 China

ISBN: (纸本)9781728163277

Learning how to model global relationships and extract local details is crucial in improving the performance of multi-organ segmentation. Most existing U-shaped structure methods use feature fusion to address these two challenges, but still lack the ability to balance capturing global relationships and local details. To address these issues, we propose a novel multi-organ segmentation framework called TransWnet to mine global relationships and local details from both intra-and inter-scale perspectives. To achieve this, we innovatively design a Row and Column Swin Transformer (RCST) module that can efficiently capture global contextual features and construct local information. Specifically, we design a parallel structure of Row and Column Attention to model the global relationships of multi-scale encoded features, and further mine local information from the global relationships through a local window mechanism. Extensive experiments on the Synapse dataset show that our method outperforms state-of-The-Art approaches and achieves accurate segmentation of abdominal multi-organs. © 2023 IEEE.

关键词： Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

UAV Visual Tracking Algorithm Based on Feature Fusion of the Attention Mechanism 5

UAV Visual Tracking Algorithm Based on Feature Fusion of the...

引用

5th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2022

作者： Ma, Sugang Zhang, Zixian Zhao, Zhixian Yang, Xiaobao Hou, Zhiqiang School of Computer Science and Technology Xi'an University of Posts and Telecommunications Shaanxi Xi'an China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications Shaanxi Xi'an China Xi'an Key Laboratory of Big Data and Intelligent Computing Xi'an University of Posts and Telecommunications Shaanxi Xi'an China

ISBN: (纸本)9781450396899

To enhance the expression ability of deep features and improve the tracking performance of the fully convolutional siamese network (SiamFC) in the UAV scene, we propose a UAV visual tracking algorithm based on feature fusion of the attention mechanism. By designing the local perception attention module and the global perception attention module to enhance the features extracted from the backbone network, a set of complementary local enhanced features and global enhanced features are obtained. And then, the tracking response map fused with the two features is then located, which effectively improves the tracking robustness of SiamFC in the UAV scene. The algorithm and nine other related algorithms such as SiamFC are tested on the DTB70 dataset. The experiments show that the algorithm has a good tracking performance and can adapt to the visual object tracking task in the UAV scene. © 2022 ACM.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Network Traffic Prediction with Attention-based Spatial-Temporal Graph Network

Network Traffic Prediction with Attention-based Spatial-Temp...

引用

IEEE Workshop on High Performance Switching and Routing

作者： Yufei Peng Yingya Guo Run Hao Junda Lin Department of Computer and Data Science Fuzhou University Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University

Network traffic prediction plays a significant role in network management. Previous network traffic prediction methods mainly focus on the temporal relationship between network traffic, and used time series models to predict network traffic, ignoring the spatial information contained in traffic data. Therefore, the prediction accuracy is limited, especially in long-term prediction. To improve the prediction accuracy of the dynamic network traffic in the long term, we propose an Attention-based Spatial-Temporal Graph Network (ASTGN) model for network traffic prediction to better capture both the temporal and spatial relations between the network traffic. Specifically, in ASTGN, we exploit an encoder-decoder architecture, where the encoder encodes the input network traffic and the decoder outputs the predicted network traffic sequences, integrating the temporal and spatial information of the network traffic data through the Spatio-Temporal Embedding module. The experimental results demonstrate the superiority of our proposed method ASTGN in long-term prediction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Simple Controllable Left-Right-Hand Circularly Polarized Antenna for GPS L2 Using Arc-like Slots

A Simple Controllable Left-Right-Hand Circularly Polarized A...

引用

IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP)

作者： Lulu Meng Sixian Qian Zhixiang Huang Yingsong Li Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University China

A single-layer, polarization adjustable circular-polarization (CP) antenna with four arc-like slots has been designed for GPS L2 band. The created antenna uses four arc-like slots to tune the phase difference to form a CP antenna, where the arc-like slots with a specific size relationship are etched on the patch. By adjusting the radius of the arc-like slots, Left-handed -circular-polarization (LHCP) and Right-handed-circular-polarization (RHCP) can be realized easily with simple structure. Simulations and optimizations show that the constructed CP-antenna has a good axial-ratio bandwidth of 10 MHz and impedance-bandwidth of 40 MHz and 30 MHz for LHCP and RHCP application.

关键词： Polarization Slot antennas Bandwidth Directive antennas Antenna feeds Optimization Global Positioning System

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：