检索结果-内蒙古大学图书馆

15th International Conference on Intelligent Computing, ICIC 2019

作者： Huang, Jin-Long Zhang, Hong-Bo Du, Ji-Xiang Zheng, Jian-Feng Peng, Xiao-Xiao Department of Computer Science and Technology Huaqiao University Xiamen China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Huaqiao University Xiamen China

ISBN: (纸本)9783030267650

Automatically product counting in the handmade process plays a vital role in the manufacturing industry, especially at the sewing industry. Nevertheless, there is currently a few methods to count the product number in the hand sewing process from surveillance video automatically. Due to the sewing procedure is a cyclical action, the product counting in hand sewing process is regarded as periodic action temporal localization and counting problem. In this paper, in order to solve this problem, we propose a novel two-path method, based on pose estimation and region-based convolutional neural network. The pose estimation method is used to obtain the trajectory information of human joint points, and the periodic action is located by detecting the periodic changes in joint trajectory. An effective two-threshold method is proposed to locate each action and count the number of periodic action from the trajectory information. To more accurately localization, we use a convolutional neural network to predict whether the workbench is empty or not. We fuse the results of joint trajectory and the status of the workbench to adjust the final periodic action localization and counting the number. To verify the proposed method, we built a new video database collected in the real sewing industry. The experimental results show that the proposed method is effective and constructive at periodic action localization and counting in the video for the sewing industry. © 2019, Springer Nature Switzerland AG.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

UformPose: A U-Shaped Hierarchical Multi-Scale keypoint-Aware Framework for Human Pose Estimation

引用

IEEE Transactions on Circuits and Systems for Video Technology 2023年第4期33卷 1697-1709页

作者： Wang, You-Jie Luo, Yan-Min Bai, Gui-Hu Guo, Jing-Ming Huaqiao University Key Laboratory for Computer Vision and Pattern Recognition The College of Computer Science and Technology Xiamen361021 China National Taiwan University of Science and Technology Department of Electrical Engineering Taipei10607 Taiwan

Human pose estimation is a fundamental yet challenging task in computer vision. However, difficult scenarios such as invisible keypoints, occlusions and small-scale persons are still not well-handed. In this paper, we present a novel pose estimation framework named UformPose which targets to relieve these issues. UformPose has two core designs: Shared Feature Pyramid Stem (SFPS) and U-shaped hierarchical Multi-scale keypoint-aware Attention Module (U-MKAM). SFPS is a feature pyramid stem with shared mechanism to learn stronger low-level features at the initial stage, and the shared mechanism can facilitate cross-resolution commonality learning. Our U-MKAM attempts to generate high-quality high-resolution representations by integrating all levels of feature representation of the backbone layer by layer. More importantly, we utilize the flexibility of attention operations for keypoint-aware modeling, which explicitly captures and trades-offs the dependencies between keypoints. We empirically demonstrate the effectiveness of our framework through the competitive pose estimation results on the COCO dataset. Extensive experiments and visual analysis on CrowdPose demonstrate the robustness of our model in crowd scenes. © 1991-2012 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Efficient Audio-Visual Speaker recognition via Deep Heterogeneous Feature Fusion 12th

Efficient Audio-Visual Speaker Recognition via Deep Heteroge...

引用

12th Chinese Conference on Biometric recognition, CCBR 2017

作者： Liu, Yu-Hang Liu, Xin Fan, Wentao Zhong, Bineng Du, Ji-Xiang Department of Computer Science Huaqiao University Xiamen361021 China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Huaqiao University Xiamen361021 China

ISBN: (纸本)9783319699226

Audio-visual speaker recognition (AVSR) has long been an active research area primarily due to its complementary information for reliable access control in biometric system, and it is a challenging problem mainly attributes to its multimodal nature. In this paper, we present an efficient audio-visual speaker recognition approach via deep heterogeneous feature fusion. First, we exploit a dual-branch deep convolutional neural networks (CNN) learning framework to extract and fuse the high-level semantic features of face and audio data. Further, by considering the temporal dependency of audio-visual data, we embed the fused features into a bidirectional Long Short-Term Memory (LSTM) networks to produce the recognition result, though which the speakers acquired under different challenging conditions can be well identified. The experimental results have demonstrated the efficiency of our proposed approach in both audio-visual feature fusion and speaker recognition. © 2017, Springer International Publishing AG.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Research on DNA sequence homology based on second order Markov model

引用

Journal of Fiber Bioengineering and Informatics 2015年第3期8卷 539-546页

作者： Zhang, Junyan Yang, Chenhui College of Computer Science Chengdu University Chengdu China Key Laboratory of Pattern Recognition and Intelligent Information Processing of Sichuan Chengdu University Chengdu China

DNA sequence homology is a critical and fundamental problem in bioinformatics. In this paper, we solve this problem by use of the second order Markov modal instead of traditional sequence alignment because DNA character sequence meets the Markov properties. Hence, the characteristics of DNA sequences are represented by using their two-step transition probabilities matrices. The similarity degree measurement between two different DNA sequences is defined. Our DSHM algorithm is put forward which is implemented by MyEclipse. The contrast experiments are done between DSHM and other two methods. The experimental results show that DSHM algorithm can determine DNA sequence homology correctly in the more effective way. © 2015 Binary Information Press & Textile Bioengineering and Informatics Society.

关键词： DNA sequences

来源：评论

学校读者我要写书评

暂无评论

A Dual-Space Framework for General Knowledge Distillation of Large Language Models

arXiv

引用

arXiv 2025年

作者： Zhang, Xue Zhang, Songming Liang, Yunlong Meng, Fandong Chen, Yufeng Xu, Jinan Zhou, Jie School of Computer Science and Technology Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence Beijing Jiaotong University Beijing100044 China Pattern Recognition Center WeChat AI Tencent Inc China

Knowledge distillation (KD) is a promising solution to compress large language models (LLMs) by transferring their knowledge to smaller models. During this process, white-box KD methods usually minimize the distance between the output distributions of the teacher model and the student model to transfer more information. However, we reveal that the current white-box KD framework exhibits two limitations: a) bridging probability distributions from different output spaces will limit the similarity between the teacher model and the student model;b) this framework cannot be applied to LLMs with different vocabularies. One of the root causes for these limitations is that the distributions from the teacher and the student for KD are output by different prediction heads, which yield distributions in different output spaces and dimensions. Therefore, in this paper, we propose a dual-space knowledge distillation (DSKD) framework that unifies the prediction heads of the teacher and the student models for KD. Specifically, we first introduce two projectors with ideal initialization to project the teacher/student hidden states into the student/teacher representation spaces. After this, the hidden states from different models can share the same head and unify the output spaces of the distributions. Furthermore, we develop an exact token alignment (ETA) algorithm to align the same tokens in two differently-tokenized sequences. Based on the above, our DSKD framework is a general KD framework that supports both off-policy and on-policy KD, and KD between any two LLMs regardless of their vocabularies. Extensive experiments on instruction-following, mathematical reasoning, and code generation benchmarks show that DSKD significantly outperforms existing methods based on the current white-box KD framework and surpasses other cross-tokenizer KD methods for LLMs with different vocabularies. © 2025, CC BY-NC-ND.

关键词： Distribution functions

来源：评论

学校读者我要写书评

暂无评论

WDA-Net: Weakly-Supervised Domain Adaptive Segmentation of Electron Microscopy

WDA-Net: Weakly-Supervised Domain Adaptive Segmentation of E...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Qiu, Dafei Yi, Jiajin Peng, Jialin Huaqiao University Xiamen Key Laboratory of Computer Vision and Pattern Recognition China Meitu Inc Xiamen China Huaqiao University College of Computer Science and Technology China

ISBN: (纸本)9781665468190

Accurate segmentation of organelle instances is essential for electron microscopy analysis. Despite the outstanding performance of fully supervised methods, they highly rely on sufficient per-pixel annotated data and are sensitive to domain shift. Aiming to develop a highly annotation-efficient approach with competitive performance, we focus on weakly-supervised domain adaptation (WDA) with a type of extremely sparse and weak annotation demanding minimal annotation efforts, i.e., sparse point annotations on only a small subset of object instances. To reduce performance degradation arising from domain shift, we explore multi-level transferable knowledge through conducting three complementary tasks, i.e., counting, detection, and segmentation, constituting a task pyramid with different levels of domain invariance. The intuition behind this is that after investigating a related source domain, it is much easier to spot similar objects in the target domain than to delineate their fine boundaries. Specifically, we enforce counting estimation as a global constraint to the detection with sparse supervision, which further guides the segmentation. A cross-position cut-and-paste augmentation is introduced to further compensate for the annotation sparsity. Extensive validations show that our model with only 15% point annotations can achieve comparable performance as supervised models and shows robustness to annotation selection. © 2022 IEEE.

关键词： Electron microscopy

来源：评论

学校读者我要写书评

暂无评论

Graph convolutional network combined with random walks and graph attention network for node classification

引用

The Journal of China Universities of Posts and Telecommunications 2024年第3期31卷 1-14页

作者： Chen Yong Xie Xiaozhu Weng Wei College of Computer and Information Engineering Xiamen University of TechnologyXiamen 361024China Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen University of TechnologyXiamen 361024China

Graph conjoint attention(CAT)network is one of the best graph convolutional networks(GCNs)frameworks,which uses a weighting mechanism to identify important neighbor ***,this weighting mechanism is learned based on static information,which means it is susceptible to noisy nodes and edges,resulting in significant *** this paper,a method is proposed to obtain context dynamically based on random walk,which allows the context-based weighting mechanism to better avoid noise ***,the proposed context-based weighting mechanism is combined with the node content-based weighting mechanism of the graph attention(GAT)network to form a model based on a mixed weighting *** model is named as the context-based and content-based graph convolutional network(CCGCN).CCGCN can better discover important neighbors,eliminate noise edges,and learn node embedding by message *** show that CCGCN achieves state-of-the-art performance on node classification tasks in multiple datasets.

关键词： graph neural network(GNN) graph convolutional network(GCN) semi-supervised classification graph analysis

来源：评论

学校读者我要写书评

暂无评论

Re-identification of Saimaa Ringed Seals from Image Sequences 23rd

Re-identification of Saimaa Ringed Seals from Image Sequen...

引用

22nd Scandinavian Conference on Image Analysis, SCIA 2023

作者： Nepovinnykh, Ekaterina Vilkman, Antti Eerola, Tuomas Kälviäinen, Heikki Computer Vision and Pattern Recognition Laboratory Department of Computational Engineering Lappeenranta-Lahti University of Technology LUT Lappeenranta Finland

ISBN: (纸本)9783031314346

Automatic game cameras are commonly used for monitoring wildlife as they allow to document of the activity of animals in a non-invasive manner. By utilizing a large number of cameras and identifying individual animals from the images, it is possible to, for example, estimate the population size and study the migration patterns of the animals. Large image volumes produced by the cameras call for automated methods for the analysis. Re-identification of animals has commonly been implemented through one-to-one matching, where images are processed individually and the best match is searched from the database of known individuals one by one. Game cameras can be configured to produce a sequence of images that allows capturing the animal from multiple angles potentially improving the re-identification accuracy. In this work, the re-identification of the endangered Saimaa ringed seal (pusa hispida saimensis) from image sequences is studied. The individual identification is realized through Saimaa ringed seal’s unique pelage pattern. The proposed one-to-many and many-to-many matching methods aggregate the pelage pattern features over the whole sequence providing better embeddings for the re-identification tasks. We show that the proposed aggregation method outperforms traditional one-to-one matching based re-identification by a large margin. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Animals

来源：评论

学校读者我要写书评

暂无评论

Smart electric valve controller based on all digital phase-locked loop

引用

Sensors and Transducers 2014年第2期165卷 89-95页

作者： Xiaoling, Li School of Computer Science and Technology Chengdu University Chengdu 610106 China University Key Laboratory of Pattern Recognition and Intelligent information Processing Chengdu 610106 China

Electric valve has been applied to various occasions and domains. In some adverse environments where such defects of traditional control system as low efficiency and poor safety have been exposed, optical encoder is adopted to detect the motor's angular speed and displacement and ADPLL is used for generating control logic, so as to regulate switching frequency and duty ratio of pulse wavelength modulation (PWM), alter the motor's input voltage and current and achieve the open and close control of Electric valve. The system can automatically regulate the rotating speed and the dragging torque of motor based on the open and close position, rotating speed and torque of the valve to make sure that the motor can achieve the best rotating speed and torque. The results of the test by motor 90ZY24-200 prove that the system has prompt response, high accuracy, little harmonic and ripple range;it can also automatically adapt to the torque alteration to realize the protective function of direct current (DC) motor. Electric valve control based on ADPLL can not only realize functions of the whole system, but also enjoys a stable and reliable performance. It can reduce the harmonic and ripple, and help the whole system achieve real-time remote control, data collection and transmission, and display the motor torque in LED, so as to control the valve flow stably and accurately through DC motor. © 2014 IFSA Publishing, S. L.

关键词： Locks (fasteners)

来源：评论

学校读者我要写书评

暂无评论

Dual Branch PnP Based Network for Monocular 6D Pose Estimation

引用

Intelligent Automation & Soft Computing 2023年第6期36卷 3243-3256页

作者： Jia-Yu Liang Hong-Bo Zhang Qing Lei Ji-Xiang Du Tian-Liang Lin Department of Computer Science and Technology Huaqiao UniversityXiamen361000China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Huaqiao UniversityXiamen361000China Fujian Key Laboratory of Big Data Intelligence and Security Huaqiao UniversityXiamen361000China College of Mechanical Engineering and Automation Xiamen361000China

Monocular 6D pose estimation is a functional task in the field of com-puter vision and *** recent years,2D-3D correspondence-based methods have achieved improved performance in multiview and depth data-based ***,for monocular 6D pose estimation,these methods are affected by the prediction results of the 2D-3D correspondences and the robustness of the per-spective-n-point(PnP)*** is still a difference in the distance from the expected estimation *** obtain a more effective feature representation result,edge enhancement is proposed to increase the shape information of the object by analyzing the influence of inaccurate 2D-3D matching on 6D pose regression and comparing the effectiveness of the intermediate ***,although the transformation matrix is composed of rotation and translation matrices from 3D model points to 2D pixel points,the two variables are essentially different and the same network cannot be used for both variables in the regression ***,to improve the effectiveness of the PnP algo-rithm,this paper designs a dual-branch PnP network to predict rotation and trans-lation ***,the proposed method is verified on the public LM,LM-O and YCB-Video *** ADD(S)values of the proposed method are 94.2 and 62.84 on the LM and LM-O datasets,*** AUC of ADD(-S)value on YCB-Video is *** experimental results show that the performance of the proposed method is superior to that of similar methods.

关键词： 6D pose monocular RGB edge enhancement dual-branch PnP 2D-3D correspondence

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：