检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Yan, Caixia Chang, Xiaojun Luo, Minnan Liu, Huan Zhang, Xiaoqin Zheng, Qinghua The Ministry of Education Key Lab of Intelligent Networks and Network Security National Engineering Lab for Big Data Analytics School of Electronic and Information Engineering Xi'an Jiaotong University Shaanxi710049 China The ReLER Lab AAAI Faculty of Engineering and Information Technology University of Technology Sydney Australia The College of Computer Science and Artificial Intelligence Wenzhou University China

Zero-shot object detection (ZSD), the task that extends conventional detection models to detecting objects from unseen categories, has emerged as a new challenge in computer vision. Most existing approaches tackle the ZSD task with a strict mapping-transfer strategy that may lead to suboptimal ZSD results: 1) the learning process of these models neglects the availab.e semantic information on unseen classes, which can easily bias towards the seen categories;2) the original visual feature space is not well-structured for the ZSD task due to the lack of discriminative information. To address these issues, we develop a novel Semantics-Guided Contrastive Network for ZSD, named ContrastZSD, a detection framework that first brings contrastive learning mechanism into the realm of zero-shot detection. Particularly, ContrastZSD incorporates two semantics-guided contrastive learning subnets that contrast between region-category and region-region pairs respectively. The pairwise contrastive tasks take advantage of supervision signals derived from both the ground truth lab.l and class similarity information. By performing supervised contrastive learning over those explicit semantic supervision, the model can learn more knowledge about unseen categories to avoid the bias problem to seen concepts, while optimizing the visual data structure to be more discriminative for better visual-semantic alignment. Extensive experiments are conducted on two popular benchmarks for ZSD, i.e., PASCAL VOC and MS COCO. Results show that our method outperforms the previous state-of-the-art on both ZSD and generalized ZSD tasks. © 2021, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Techniques for Automatic Lateral X-ray Cephalometric Landmark Detection: Is the Problem Solved?

arXiv

引用

arXiv 2024年

作者： Zhang, Hongyuan Wang, Ching-Wei Muzakky, Hikam Dai, Juan Li, Xuguang Ma, Chenglong Wu, Qian Cui, Xianan Xu, Kunlun He, Pengfei Guo, Dongqian Wang, Xianlong Lee, Hyunseok Zhong, Zhangnan Zhu, Zhu Huang, Bingsheng Medical AI Lab School of Biomedical Engineering Shenzhen University Medical School Shenzhen University Shenzhen518060 China Guangdong Key Laboratory of Biomedical Measurements and Ultrasound Imaging School of Biomedical Engineering Shenzhen University Medical School Shenzhen University Shenzhen518060 China Graduate Institute of Biomedical Engineering National Taiwan University of Science and Technology Taiwan Department of Stomatology Shenzhen University General Hospital Shenzhen University Shenzhen518055 China Institute of Stomatological Research Shenzhen University Shenzhen518055 China Chohotech Technology Co. Ltd. Lianchuang Street Yuhang District Hangzhou311422 China Singapore School of Computer Science Wuhan University Wuhan430072 China Wangxuan Institute of Computer Technology Peking University Beijing100871 China School of Artificial Intelligence Xidian University Xi’an710126 China State Key Laboratory of Internet of Things for Smart City University of Macau Taipa China The Faculty of Information Science and Engineering Ocean University of China Qingdao266100 China Digital Healthcare Department Daegu-Gyeongbuk Medical Innovation Foundation Daegu427724 Korea Republic of School of Life Science South China Normal University Guangzhou510631 China Information and Data Department Children’s Hospital Zhejiang University School of Medicine Hangzhou310000 China

Localization of the craniofacial landmarks from lateral cephalograms is a fundamental task in cephalometric analysis. The automation of the corresponding tasks has thus been the subject of intense research over the past decades. In this paper, we introduce the "Cephalometric Landmark Detection (CL-Detection)" dataset, which is the largest publicly availab.e and comprehensive dataset for cephalometric landmark detection. This multi-center and multi-vendor dataset includes 600 lateral X-ray images with 38 landmarks acquired with different equipment from three medical centers. The overarching objective of this paper is to measure how far state-of-the-art deep learning methods can go for cephalometric landmark detection. Following the 2023 MICCAI CL-Detection Challenge, we report the results of the top ten research groups using deep learning methods. Results show that the best methods closely approximate the expert analysis, achieving a mean detection rate of 75.719% and a mean radial error of 1.518 mm. While there is room for improvement, these findings undeniably open the door to highly accurate and fully automatic location of craniofacial landmarks. We also identify scenarios for which deep learning methods are still failing. Both the dataset and detailed results are publicly availab.e online, while the platform will remain open for the community to benchmark future algorithm developments at https: //***/. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Reversible data hiding in JPEG images based on multiple histograms modification 5th

Reversible data hiding in JPEG images based on multiple hist...

引用

5th International Conference on Artificial Intelligence and security, ICAIS 2019

作者： Lu, Xing Huang, Fangjun Kim, Hyoung Joong Guangdong Provincial Key Laboratory of Information Security Technology School of Data and Computer Science Sun Yat-sen University GuangzhouGuangdong510006 China Graduate School of Information Security Korea University Seoul136-701 Korea Republic of

ISBN: (纸本)9783030242701

The joint photographic experts group (JPEG) is the most popular image on the internet nowadays. Therefore, reversible data hiding (RDH), which can be used to ensure the widely used JPEG images' originality and integrity, has attracted more and more attention. However, when performing RDH in JPEG images, besides the embedding capacity and visual quality, which have to be considered for uncompressed images, the storage size of the marked JPEG file should also be considered. In this paper, a new JPEG RDH scheme based on multiple histograms modification is proposed. Firstly, according to zero run length (ZRL) of the discrete cosine transform (DCT) coefficients, multiple DCT coefficient histograms are generated. Then, the message bits are embedded via the histogram shifting algorithm. In order to improve the efficiency of the proposed method, a new adaptive embedding strategy based on the ZRL and the number of zero coefficients in each 8 × 8 block is also proposed. Experimental results show that our method can effectively reduce the increase of image storage size caused by information embedding. © Springer Nature Switzerland AG 2019.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

An improved exact sampling algorithm for the standard normal distribution

arXiv

引用

arXiv 2020年

作者： Du, Yusong Fan, Baoying Wei, Baodian School of Data and Computer Science Sun Yat-sen University Guangzhou510006 China Guangdong Key Laboratory of Information Security Technology Guangzhou510006 China

In 2016, Karney proposed an exact sampling algorithm for the standard normal distribution. In this paper, we study the computational complexity of this algorithm under the random deviate model. Specifically, Karney’s algorithm requires the access to an infinite sequence of independently and uniformly random deviates over the range (0, 1). We give an estimate of the expected number of uniform deviates used by this algorithm until outputting a sample value, and present an improved algorithm with lower uniform deviate consumption. The experimental results also shows that our improved algorithm has better performance than Karney’s *** Codes 65C10 Copyright © 2020, The Authors. All rights reserved.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Dense-captioning events in videos: Sysu submission to Activitynet Challenge 2020

arXiv

引用

arXiv 2020年

作者： Wang, Teng Zheng, Huicheng Yu, Mingjing School of Data and Computer Science Sun Yat-sen University Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangdong Province Key Laboratory of Information Security Technology China

This technical report presents a brief description of our submission to the dense video captioning task of ActivityNet Challenge 2020. Our approach follows a two-stage pipeline: first, we extract a set of temporal event proposals;then we propose a multi-event captioning model to capture the event-level temporal relationships and effectively fuse the multi-modal information. Our approach achieves a 9.28 METEOR score on the test set. Copyright © 2020, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Block Compressed Sensing-Based Joint Source-Channel Coding for Wireless Image Transmission

Block Compressed Sensing-Based Joint Source-Channel Coding f...

引用

International Conference on Wireless Communications and Signal Processing (WCSP)

作者： Leqi Zheng Tingting Zhu Xiao Ma School of Data and Computer Science Sun Yat-sen University Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

ISBN: (数字)9781728172361

ISBN: (纸本)9781728172378

This paper presents a new joint source-channel coding (JSCC) scheme for wireless image transmission, which combines block compressed sensing (BCS) with systematic convolutional low density generator matrix (SC-LDGM) codes. In the proposed scheme (referred to as JSCC with BCS and SC-LDGM), an original image is firstly compressed by BCS and then quantized by dead-zone quantizer into ternary symbols. The resulting quantized sequence is encoded directly by a ternary SC-LDGM code and modulated by ternary pulse amplitude modulation (3-PAM). This scheme has the advantages to avoid the use of entropy coding and to match the quantized symbols with ternary coded modulations. Simulation results show that, the proposed scheme outperforms the conventional unequal error protection (UEP) scheme about 6 dB in terms of the energy-to-noise ratio (ENR).

关键词： Image coding Image reconstruction Wireless communication Image communication Compressed sensing Wireless sensor networks Channel coding

来源：评论

学校读者我要写书评

暂无评论

Universal Physical Camouflage Attacks on Object Detectors

Universal Physical Camouflage Attacks on Object Detectors

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Lifeng Huang Chengying Gao Yuyin Zhou Cihang Xie Alan L. Yuille Changqing Zou Ning Liu School of Data and Computer Science Sun Yat-sen University Guangdong Key Laboratory of Information Security Technology Department of Computer Science The Johns Hopkins University Max Planck Institute for Informatics University of Maryland College Park

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

In this paper, we study physical adversarial attacks on object detectors in the wild. Previous works mostly craft instance-dependent perturbations only for rigid or planar objects. To this end, we propose to learn an adversarial pattern to effectively attack all instances belonging to the same object category, referred to as Universal Physical Camouflage Attack (UPC). Concretely, UPC crafts camouflage by jointly fooling the region proposal network, as well as misleading the classifier and the regressor to output errors. In order to make UPC effective for non-rigid or non-planar objects, we introduce a set of transformations for mimicking deformable properties. We additionally impose optimization constraint to make generated patterns look natural to human observers. To fairly evaluate the effectiveness of different physical-world attacks, we present the first standardized virtual database, AttackScenes, which simulates the real 3D world in a controllab.e and reproducible environment. Extensive experiments suggest the superiority of our proposed UPC compared with existing physical adversarial attackers not only in virtual environments (AttackScenes), but also in real-world physical environments.

关键词： Proposals Detectors Semantics Perturbation methods Strain Optimization Three-dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Dynamic NOMA-Based Computation Offloading in Vehicular Platoons

arXiv

引用

arXiv 2021年

作者： Zheng, Dongsheng Chen, Yingyang Wei, Lai Jiao, Bingli Hanzo, Lajos The Department of Electronics Peking University Beijing100871 China The College of Information Science and Technology Jinan University Guangzhou510632 China The Guangdong Provincial Key Laboratory of Data Security and Privacy Protection Jinan University Guangzhou510632 China The School of Electronics and Computer Science University of Southampton SouthamptonSO17 1BJ United Kingdom

Both the Mobile edge computing (MEC)-based and fog computing (FC)-aided Internet of Vehicles (IoV) constitute promising paradigms of meeting the demands of low-latency pervasive computing. To this end, we construct a dynamic NOMA-based computation offloading scheme for vehicular platoons on highways, where the vehicles can offload their computing tasks to other platoon members. To cope with the rapidly fluctuating channel quality, we divide the timeline into successive time slots according to the channel's coherence time. Robust computing and offloading decisions are made for each time slot after taking the channel estimation errors into account. Considering a certain time slot, we first analytically characterize both the locally computed source data and the offloaded source data as well as the energy consumption of every vehicle in the platoons. We then formulate the problem of minimizing the long-term maximum task queue by optimizing the allocation of both the communication and computing resources. To solve the problem formulated, we design an online algorithm based on the classic Lyapunov optimization method and successive convex approximation (SCA) method. Finally, the numerical simulation results characterize the performance of our algorithm and demonstrate its advantages both over the local computing scheme and the orthogonal multiple access (OMA)-based offloading scheme. © 2021, CC BY.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

Cross-modal consensus network for weakly supervised temporal action localization

arXiv

引用

arXiv 2021年

作者： Hong, Fa-Ting Feng, Jia-Chang Xu, Dan Shan, Ying Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Peng Cheng Laboratory Shenzhen China Tencent PCG Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Department of Computer Science and Engineering HKUST Hong Kong Pazhou Lab Guangzhou China Guangdong Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

Weakly supervised temporal action localization (WS-TAL) is a challenging task that aims to localize action instances in the given video with video-level categorical supervision. Previous works use the appearance and motion features extracted from pre-trained feature encoder directly, e.g., feature concatenation or score-level fusion. In this work, we argue that the features extracted from the pre-trained extractors, e.g., I3D, which are trained for trimmed video action classification, but not specific for WS-TAL task, leading to inevitable redundancy and sub-optimization. Therefore, the feature re-calibration is needed for reducing the task-irrelevant information redundancy. Here, we propose a cross-modal consensus network (CO2-Net) to tackle this problem. In CO2-Net, we mainly introduce two identical proposed cross-modal consensus modules (CCM) that design a cross-modal attention mechanism to filter out the task-irrelevant information redundancy using the global information from the main modality and the cross-modal local information from the auxiliary modality. Moreover, we further explore inter-modality consistency, where we treat the attention weights derived from each CCM as the pseudo targets of the attention weights derived from another CCM to maintain the consistency between the predictions derived from two CCMs, forming a mutual learning manner. Finally, we conduct extensive experiments on two commonly used temporal action localization datasets, THUMOS14 and ActivityNet1.2, to verify our method, which we achieve the state-of-the-art results. The experimental results show that our proposed cross-modal consensus module can produce more representative features for temporal action localization. Copyright © 2021, The Authors. All rights reserved.

关键词： Carbon dioxide

来源：评论

学校读者我要写书评

暂无评论

Improving fast segmentation with teacher-student learning 29

Improving fast segmentation with teacher-student learning

引用

29th British Machine Vision Conference, BMVC 2018

作者： Xie, Jiafeng Shuai, Bing Hu, Jian-Fang Lin, Jingyang Zheng, Wei-Shi School of Data and Computer Science Sun Yat-sen University China Nanyang Technological University Singapore Guangdong Key Laboratory of Information Security Technology China Key Laboratory of Machine Intelligence and Advanced Computing MOE Hong Kong

Recently, segmentation neural networks have been significantly improved by demonstrating very promising accuracies on public benchmarks. However, these models are very heavy and generally suffer from low inference speed, which limits their application scenarios in practice. Meanwhile, existing fast segmentation models usually fail to obtain satisfactory segmentation accuracies on public benchmarks. In this paper, we propose a teacher-student learning framework that transfers the knowledge gained by a heavy and better performed segmentation network (i.e. teacher) to guide the learning of fast segmentation networks (i.e. student). Specifically, both zero-order and first-order knowledge depicted in the fine annotated images and unlab.led auxiliary data are transferred to regularize our student learning. The proposed method can improve existing fast segmentation models without incurring extra computational overhead, so it can still process images with the same fast speed. Extensive experiments on the Pascal Context, Cityscape and VOC 2012 datasets demonstrate that the proposed teacher-student learning framework is able to significantly boost the performance of student network. © 2018. The copyright of this document resides with its authors.

关键词： Students

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：