检索结果-内蒙古大学图书馆

Large-Scale Video-Based Person Re-identification via Non-local Attention and Feature Erasing 16th

学校读者我要写书评

暂无评论

Large-Scale Video-Based Person Re-identification via Non-loc...

16th International Forum on Digital TV and Wireless Multimedia communication, IFTC 2019

作者： Yang, Zhao Chang, Zhigang Zheng, Shibao Institute of Image Communication and Network Engineering Shanghai Key Labs of Digital Media Processing and Transmission Shanghai Jiao Tong University SEIEE Buildings 5-421 Shanghai200240 China

ISBN: (纸本)9789811533402

Encoding the video tracks of person to an aggregative representation is the key for video-based person re-identification (re-ID), where average pooling or RNN methods are typically used to aggregating frame-level features. However, It is still difficult to deal with the spatial misalignment caused by occlusion, posture changes and camera views. Inspired by the success of non-local block in video analysis, we use a non-local attention block as a spatial-temporal attention mechanism to handle the spatial-temporal misalignment problem. Moreover, partial occlusion is widely occurred in video sequences. We propose a local feature branch to tackle the partial occlusion problem by using feature erasing in the frame-level feature map. Therefore, our network is composed by two-branch, the global branch via non-local attention encoding the global feature and the local feature branch grasping the local feature. In evaluation, the global feature and local feature are concatenated to obtain a more discriminative feature. We conduct extensive experiments on two challenging datasets (MARS and iLIDS-VID). The experimental results demonstrate that our method is comparable with the state-of-the-art methods in these datasets. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Signal encoding

A No-Reference Quality Assessment Method for Digital Human Head

学校读者我要写书评

暂无评论

A No-Reference Quality Assessment Method for Digital Human H...

IEEE International Conference on image processing

作者： Yingjie Zhou Zicheng Zhang Wei Sun Xiongkuo Min Xianghe Ma Guangtao Zhai Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China School of Information and Control Engineering China University of Mining Technology China Peng Cheng Laboratory China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China

In recent years, digital humans have been widely applied in augmented/virtual reality (A/VR), where viewers are allowed to freely observe and interact with the volumetric content. However, the digital humans may be degraded with various distortions during the procedure of generation and transmission. Moreover, little effort has been put into the perceptual quality assessment of digital humans. Therefore, it is urgent to carry out objective quality assessment methods to tackle the challenge of digital human quality assessment (DHQA). In this paper, we develop a novel no-reference (NR) method based on Transformer to deal with DHQA in a multi-task manner. Specifically, the front 2D projections of the digital humans are rendered as inputs and the vision transformer (ViT) is employed for the feature extraction. Then we design a multi-task module to jointly classify the distortion types and predict the perceptual quality levels of digital humans. The experimental results show that the proposed method well correlates with the subjective ratings and outperforms the state-of-the-art quality assessment methods.

关键词：

MM-PCQA: Multi-Modal Learning for No-reference Point Cloud Quality Assessment

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Zicheng Sun, Wei Min, Xiongkuo Wang, Qiyuan He, Jun Zhou, Quan Zhai, Guangtao Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China Bilibili Inc Shanghai China

The visual quality of point clouds has been greatly emphasized since the ever-increasing 3D vision applications are expected to provide cost-effective and high-quality experiences for users. Looking back on the development of point cloud quality assessment (PCQA), the visual quality is usually evaluated by utilizing single-modal information, i.e., either extracted from the 2D projections or 3D point cloud. The 2D projections contain rich texture and semantic information but are highly dependent on viewpoints, while the 3D point clouds are more sensitive to geometry distortions and invariant to viewpoints. Therefore, to leverage the advantages of both point cloud and projected image modalities, we propose a novel no-reference MultiModal Point Cloud Quality Assessment (MM-PCQA) metric. In specific, we split the point clouds into sub-models to represent local geometry distortions such as point shift and down-sampling. Then we render the point clouds into 2D image projections for texture feature extraction. To achieve the goals, the sub-models and projected images are encoded with point-based and image-based neural networks. Finally, symmetric cross-modal attention is employed to fuse multi-modal quality-aware information. Experimental results show that our approach outperforms all compared state-of-the-art methods and is far ahead of previous no-reference PCQA methods, which highlights the effectiveness of the proposed method. The code is available at https://***/zzc-1998/MM-PCQA. © 2022, CC BY.

关键词： Cost effectiveness

A NO-REFERENCE QUALITY ASSESSMENT METHOD FOR DIGITAL HUMAN HEAD

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhou, Yingjie Zhang, Zicheng Sun, Wei Min, Xiongkuo Ma, Xianghe Zhai, Guangtao Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China School of Information and Control Engineering China University of Mining Technology China Peng Cheng Laboratory China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China

关键词： Textures

3D-ANAS: 3D Asymmetric neural architecture search for fast Hyperspectral image classification

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Haokui Gong, Chengrong Bai, Yunpeng Bai, Zongwen Li, Ying the Shaanxi Provincial Key Lab of Speech and Image Information Processing School of Computer Science Northwestern Polytechnical University Xi’an710129 China

Hyperspectral images involve abundant spectral and spatial information, playing an irreplaceable role in land-cover classification. Recently, based on deep learning technologies, an increasing number of HSI classification approaches have been proposed, which demonstrate promising performance. However, previous studies suffer from two major drawbacks: 1) the architecture of most deep learning models is manually designed, relies on specialized knowledge, and is relatively tedious. Moreover, in HSI classifications, datasets captured by different sensors have different physical properties. Correspondingly, different models need to be designed for different datasets, which further increases the workload of designing architectures;2) the mainstream framework is a patch-to-pixel framework. The overlap regions of patches of adjacent pixels are calculated repeatedly, which increases computational cost and time cost. Besides, the classification accuracy is sensitive to the patch size, which is artificially set based on extensive investigation experiments. To overcome the issues mentioned above, we firstly propose a 3D asymmetric neural network search algorithm and leverage it to automatically search for efficient architectures for HSI classifications. By analysing the characteristics of HSIs, we specifically build a 3D asymmetric decomposition search space, where spectral and spatial information are processed with different decomposition convolutions. Furthermore, we propose a new fast classification framework, i,e., pixel-to-pixel classification framework, which has no repetitive operations and reduces the overall cost. Experiments on three public HSI datasets captured by different sensors demonstrate the networks designed by our 3D-ANAS achieve competitive performance compared to several state-of-the-art methods, while having a much faster inference speed. Code is available at: https://***/hkzhang91/3D-ANAS. © 2021, CC BY.

关键词： Pixels

An Improved Epsilon Method with M2M for Solving Imbalanced CMOPs with Simultaneous Convergence-Hard and Diversity-Hard Constraints 11th

学校读者我要写书评

暂无评论

An Improved Epsilon Method with M2M for Solving Imbalanced C...

11th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2021

作者： Fan, Zhun Yang, Zhi Tang, Yajuan Li, Wenji Xu, Biao Wang, Zhaojun Sun, Fuzan Long, Zhoubin Zhu, Guijie Department of Electronic and Information Engineering Shantou University Shantou Guangdong515063 China Key Lab of Digital Signal and Image Processing of Guangdong Province Shantou University Shantou Guangdong515063 China Ministry of Education Shantou Guangdong515063 China State Key Lab of Digital Manufacturing Equipment and Technology Huazhong University of Science and Technology Wuhan430000 China

ISBN: (纸本)9783030720612

When tackling imbalanced constrained multi-objective optimization problems (CMOPs) with simultaneous convergence-hard and diversity-hard constraints, a critical issue is to balance the diversity and convergence of populations. To address this issue, this paper proposes a hybrid algorithm which combines an improved epsilon constraint-handling method (IEpsilon) with a multi-objective to multi-objective (M2M) decomposition approach, namely M2M-IEpsilon. The M2M decomposition mechanism in M2M-IEpislon has the capability to deal with imbalanced objectives. The IEpsilon constraint-handling method can prevent populations falling into large infeasible regions, thus improves the convergence performance of the proposed algorithm. To verify the performance of the proposed M2M-IEpsilon, a series of imbalanced CMOPs with simultaneous convergence-hard and diversity-hard constraints, namely ICD-CMOPs, is designed by using the DAS-CMOPs framework. Six state-of-the-art constrained multi-objective evolutionary algorithms (CMOEAs), including CM2M, CM2M2, NSGA-II-CDP, MOEA/D-CDP, MOEA/D-IEpsilon and PPS-MOEA/D, are employed to compare with M2M-IEpsilon on the ICD-CMOPs. Through the analysis of experimental results, the proposed M2M-IEpsilon is superior to the other six algorithms in solving ICD-CMOPs, which illustrates the superiority of the proposed M2M-IEpsilon in dealing with ICD-CMOPs with simultaneous convergence-hard and diversity-hard constraints. © 2021, Springer Nature Switzerland AG.

关键词： Constraint handling

A Pilot Study of Relating MYCN-Gene Amplification with Neuroblastoma-Patient CT Scans

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Zihan Xiang, Xiang Peng, Xuehua Shao, Jianbo MoE Key Lab of Image Info Processing & Intelligent Control School of Artificial Intelligence & Automation China Wuhan Children's Hospital Tongji Medical College China Huazhong University of Science and Technology Wuhan430074 China

Neuroblastoma is one of the most common cancers in infants, and the initial diagnosis of this disease is difficult. At present, the MYCN gene amplification (MNA) status is detected by invasive pathological examination of tumor samples. This is time-consuming and may have a hidden impact on children. To handle this problem, we adopt multiple machine learning (ML) algorithms to predict the presence or absence of MYCN gene amplification. The dataset is composed of retrospective CT images of 23 neuroblastoma patients. Different from previous work, we develop the algorithm without manually-segmented primary tumors which is time-consuming and not practical. Instead, we only need the coordinate of the center point and the number of tumor slices given by a subspecialty-trained pediatric radiologist. Specifically, CNN-based method uses pre-trained convolutional neural network, and radiomics-based method extracts radiomics features. Our results show that CNN-based method outperforms the radiomics-based method. Copyright © 2022, The Authors. All rights reserved.

关键词： Computerized tomography

Indirect effects among biodiversity loss of mutualistic ecosystems

学校读者我要写书评

暂无评论

National Science Open 2022年第2期1卷 8-18页

作者： Guangwei Wang Xueming Liu Guanrong Chen Hai-Tao Zhang School of Artificial Intelligence and Automation Key Laboratory of Image Processing and Intelligent ControlHuazhong University of Science and TechnologyWuhan 430074China State Key Lab of Digital Manufacturing Equipment and Technology Huazhong University of Science and TechnologyWuhan 430074China Department of Electrical Engineering City University of Hong KongHong KongChina

Drastic reduction in biodiversity has been a severe threat to ecosystems,which is exacerbated when losing few species leads to disastrous and even irreparable ***,revealing the mechanism underlying biodiversity loss is of uttermost *** this study,we show that abundant indirect interactions among mutualistic ecosystems are critical in determining species’*** topological and ecological characteristics,we propose an indicator derived from a dynamic model to identify keystone species and quantify their influence,which outperforms widely-used indicators like degree in realistic and simulated ***,we demonstrate that networks with high modularity,heterogeneity,biodiversity,and less intimate interactions tend to have larger indirect effects,which are more amenable in predicting decline of biodiversity with the proposed *** findings shed some light onto the influence of apposite biodiversities,paving the way from complex network theory to ecosystem protection and restoration.

关键词： biodiversity indirect effect complex network mutualism

Design of three-dimensional gesture recognition and motion tracking human-computer intelligent interaction system based on PAJ7620

学校读者我要写书评

暂无评论

Design of three-dimensional gesture recognition and motion t...

2021 International Conference on Information Technology and Intelligent Control, CITIC 2021

作者： Wang, Jian-Liang Wei, Ye Li, Yang Meng, Yuan Li, Ming-Yu Xu, Jiang-Yan Jiao, Wan-Ming Ban, Lan Liu, Jun-Pei Zhang, Hong-Feng Zhang, Wei Wang, Li-Na Luo, Fu-Chen Wang, Yu-Duo Tianjin College University of Science and Technology Beijing Tianjin China Civil Aviation University of China Tianjin Key Laboratory of Intelligent Signal and Image Processing Tianjin China Tianjin Normal University Tianjin Key Laboratory of Wireless Mobile Communication and Radio Energy Transmission Tianjin China Binhai College Nankai University Tianjin China Tianjin Tianshi College Tianjin China

With the increasing degree of informatization and intelligence in modern society, human-computer interaction technology has become one of the commonly used scientific and technological achievements. "Gesture" can be used as inputing image information, and make more intuitive and convenient for perception of user interface. To a certain extent, it can complete the intelligent interaction function between human and computer. The design is a human-computer intelligent interaction system for 3D gesture recognition and motion tracking based on gesture sensor PAJ7620. STM32 chip is used as the main control unit to recognize accurately, analyze and process different gestures collected by the gesture sensor, and control the OLED screen to display the gesture information. It can further drive the carrier to take corresponding actions and realize the intelligent human-computer interaction recognition of gestures. The experimental test results show that the recognition rate is lose to 100% when the distance is within 10cm. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Gesture recognition