检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Dong, Zhangfu He, Yuting Qi, Xiaoming Chen, Yang Shu, Huazhong Coatrieux, Jean-Louis Yang, Guanyu Li, Shuo LIST Key Laboratory of Computer Network and Information Integration [Southeast University Ministry of Education Nanjing China Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing China Dept. of Medical Biophysics University of Western Ontario LondonON Canada

The nature of thick-slice scanning causes severe inter-slice discontinuities of 3D medical images, and the vanilla 2D/3D convolutional neural networks (CNNs) fail to represent sparse inter-slice information and dense intra-slice information in a balanced way, leading to severe underfitting to inter-slice features (for vanilla 2D CNNs) and over-fitting to noise from long-range slices (for vanilla 3D CNNs). In this work, a novel mesh network (MNet) is proposed to balance the spatial representation inter axes via learning. 1) Our MNet latently fuses plenty of representation processes by embedding multi-dimensional convolutions deeply into basic modules, making the selections of representation processes flexible, thus balancing representation for sparse inter-slice information and dense intra-slice information adaptively. 2) Our MNet latently fuses multi-dimensional features inside each basic module, simultaneously taking the advantages of 2D (high segmentation accuracy of the easily recognized regions in 2D view) and 3D (high smoothness of 3D organ contour) representations, thus obtaining more accurate modeling for target regions. Comprehensive experiments are performed on four public datasets (CT&MR), the results consistently demonstrate the proposed MNet outperforms the other methods. The code and datasets are available at: https://***/zfdong-code/MNet. © 2022, CC BY-NC-ND.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Visible-Thermal Multiple Object Tracking: Large-scale Video Dataset and Progressive Fusion Approach

arXiv

引用

arXiv 2024年

作者： Zhu, Yabin Wang, Qianwu Li, Chenglong Tang, Jin Huang, Zhixiang The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University School of Public Safety and Emergency Management Anhui University of Science and Technology Hefei231131 China School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Security Artificial Intelligence School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China Center for Big Data and Population Health of IHM China

The complementary benefits from visible and thermal infrared data are widely utilized in various computer vision task, such as visual tracking, semantic segmentation and object detection, but rarely explored in Multiple Object Tracking (MOT). In this work, we contribute a large-scale Visible-Thermal video benchmark for MOT, called VT-MOT. VT-MOT has the following main advantages. 1) The data is large scale and high diversity. VT-MOT includes 582 video sequence pairs, 401k frame pairs from surveillance, drone, and handheld platforms. 2) The cross-modal alignment is highly accurate. We invite several professionals to perform both spatial and temporal alignment frame by frame. 3) The annotation is dense and high-quality. VT-MOT has 3.99 million annotation boxes annotated and double-checked by professionals, including heavy occlusion and object re-acquisition (object disappear and reappear) challenges. To provide a strong baseline, we design a simple yet effective tracking framework, which effectively fuses temporal information and complementary information of two modalities in a progressive manner, for robust visible-thermal MOT. A comprehensive experiment are conducted on VT-MOT and the results prove the superiority and effectiveness of the proposed method compared with state-of-the-art methods. From the evaluation results and analysis, we specify several potential future directions for visible-thermal MOT. The project is released in https://***/wqw123wqw/PFTrack. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Manifold Learning Inspired Mating Restriction for Evolutionary Constrained Multiobjective Optimization 11th

Manifold Learning Inspired Mating Restriction for Evolutiona...

引用

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

作者： Li, Lianghao He, Cheng Cheng, Ran Pan, Linqiang Key Laboratory of Image Information Processing and Intelligent Control of Education Ministry of China School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China

ISBN: (纸本)9783030720612

Mating restriction strategies are capable of restricting the distribution of parent solutions for effective offspring generation in evolutionary algorithms (EAs). Studies have shown the importance of these strategies in improving the performance of EAs for multiobjective optimization. Our previous study proposed a specific manifold learning inspired mating restriction (MLMR) strategy. It has shown promising capability of solving multiobjective optimization problems (MOPs) with complicated Pareto set shapes. However, the effect of mating restriction strategies in solving constrained MOPs is yet to be well studied. Here, we investigate the effectiveness of MLMR for solving constrained MOPs. The MLMR strategy is embedded into some representative multiobjective EAs and tested on various benchmark constrained MOPs. Experimental results indicate the encouraging performance of MLMR in constrained multiobjective optimization. © 2021, Springer Nature Switzerland AG.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

High-Altitude Pedestrian Detection Based on Improved YOLOv3 1

引用

6th International Conference on Data Mining and Big Data, DMBD 2021

作者： Tian, Qing Cao, Pengfei Fan, Haoyi Hu, Rong Li, Zuoyong School of Computer Science and Mathematics Fujian University of Technology Fuzhou350118 China University of Sevilla Sevilla Spain School of Computer Science and Technology Harbin University of Science and Technology Harbin150080 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Control Engineering Minjiang University Fuzhou350121 China

ISBN: (数字)9789811674761

ISBN: (纸本)9789811674754

As one of the main tasks in the field of computer vision, pedestrian detection aims to find out all pedestrians in the image or video. The existing YOLOv3 is a relatively mature object detection method. However, for the long-distance pedestrian detection task in high-altitude scenes, YOLOv3 has the limitations of low detection speed and low detection accuracy. This paper proposes an improved YOLOv3 method briefly called YOLOv3-M for the high-altitude pedestrian detection, which replaces the feature extraction module called darknet53 in YOLOv3 with MobileNetv1. Specifically, YOLOv3-M first constructs the dataset with the small objects of high-altitude pedestrians as the detection object. Then, it uses the K-means + + algorithm to re-cluster the high-altitude pedestrian dataset. Next, it uses the Distance Intersection over Union (DIoU) loss function to alleviate the problem of high-altitude pedestrian overlapping. Experimental results show that the proposed YOLOv3-M improves the detection precision and the detection speed compared to YOLOv3. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Expert-level vision-language foundation model for real-world radiology and comprehensive evaluation

arXiv

引用

arXiv 2024年

作者： Liu, Xiaohong Yang, Guoxing Luo, Yulin Mao, Jiaji Zhang, Xiang Gao, Ming Zhang, Shanghang Shen, Jun Wang, Guangyu Department of Radiology Sun Yat-Sen Memorial Hospital Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation Medical Research Center Sun Yat-Sen Memorial Hospital Sun Yat-Sen University Guangzhou China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University Beijing China

Radiology is a vital and complex component of modern clinical workflow and covers many tasks. Recently, vision-language (VL) foundation models in medicine have shown potential in processing multimodal information, offering a unified solution for various radiology tasks. However, existing studies either pre-trained VL models on natural data or did not fully integrate vision-language architecture and pretraining, often neglecting the unique multimodal complexity in radiology images and their textual contexts. Additionally, their practical applicability in real-world scenarios remains underexplored. Here, we present RadFound, a large and open-source vision-language foundation model tailored for radiology, that is trained on the most extensive dataset of over 8.1 million images and 250,000 image-text pairs, covering 19 major organ systems and 10 imaging modalities. To establish expert-level multimodal perception and generation capabilities, RadFound introduces an enhanced vision encoder to capture intra-image local features and inter-image contextual information, and a unified cross-modal learning design tailored to radiology. To fully assess the models' capability, we construct a benchmark, RadVLBench, including radiology interpretation tasks like medical vision-language question-answering, as well as text generation tasks ranging from captioning to report generation. We also propose a human evaluation framework. When evaluated on the real-world benchmark involving three representative modalities, 2D images (chest X-rays), multi-view images (mammograms), and 3D images (thyroid CT scans), RadFound significantly outperforms other VL foundation models on both quantitative metrics and human evaluation. In summary, the development of RadFound represents an advancement in radiology generalists, demonstrating broad applicability potential for integration into clinical workflows. © 2024, CC BY.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

On the Generalization of Machine-Learning-aided QoT Estimation in Optical Networks

On the Generalization of Machine-Learning-aided QoT Estimati...

引用

International Conference on Wireless and Optical Communications ( WOCC)

作者： Hanyu Gao Liang Zhang Bin Zhang Xiaoliang Chen Zhaohui Li Guangdong Provincial Key Labratory of Optoelectronic Information Processing Chips and Systems School of Electronics and Information Technology Sun Yat-sen University Guangzhou China Southern Marine Science and Engineering Guangdong Laboratory Zhuhai China Department of Electrical and Computer Engineering George Mason University Fairfax VA USA

This paper presents a composable machine learning method for generalizing the quality-of-transmission (QoT) metric estimation in optical networks. The composable machine learning approach characterizes this metric for lightpaths of arbitrary lengths by compositions of launch, propagation and readout modules. Results verify the feasibility of the design and show its successful application in facilitating autonomous lightpath provisioning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Charging Out Rectangle: Quadrangle IoU for Scene Text Detection under Balanced Supervision 25

Charging Out Rectangle: Quadrangle IoU for Scene Text Detect...

引用

Proceedings of the 4th International Conference on computer, Artificial Intelligence and Control Engineering

作者： Min Ma Xiangnan Zhang Wang Bin Qiuli Wang Kexing Che School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing Jiangsu China Key Laboratory of Intelligent Information Processing Corporation Key Laboratory of Aperture Array and Space Application The 38th Research Institute of China Electronics Technology Group Hefei Anhui China Government and Enterprise Division China Unicom Shandong Branch Jinan Shandong China School of Electronic Engineering Xidian University Xi' an Shaanxi China

ISBN: (纸本)9798400712647

In the past few years, text detection approaches have been able to cope with more and more challenging complex scenes. In this field, the arbitrary shape and various scale of text instances are the two most frequently mentioned topics. Among the existing methods, the bounding box prediction methods based on Intersection over Union (IoU) are efficient, but can only handle multi-oriented rectangles. Although arbitrary shape text detection methods have been proposed, these methods are often aimed at curved text. Besides, the post-processing is usually cumbersome. Considering that the affine transformation phenomenon caused by out-plane rotation is more frequent than curved text in real scenario, we design a quadrangle IoU (QIoU) loss to efficiently predict arbitrary convex quadrangle text. Furthermore, we analyze the challenge of variable scale in text detection from the perspective of unbalanced supervision information, and propose a cross-scale knowledge distillation (CSKD) method to improve the detection accuracy of large-scale text. An extensive ablation study demonstrates the effectiveness and superiority of the proposed strategies. During inferring, our method does not consume extra time. Using the basic feature pyramid network for feature extraction, our method achieves state-of-the-art results on several popular datasets for scene text detection.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DAM: Degradation-aware Model for Ultrasound Image Quality Assessment

引用

IEEE Journal of Biomedical and Health Informatics 2025年 PP卷 PP页

作者： Liu, Tuo Zhang, Xuan Ma, Xiuzhu Chen, Shuang Wang, Xuejuan Zhou, Ping Chen, Yang Zhou, Guangquan Lv, Faqin Southeast University School of Biological Science and Medical Engineering Nanjing 211189 China Chinese People's Liberation Army General Hospital Medical Center 3 Beijing 100039 China Southeast University Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing the Laboratory of Image Science and Technology the School of Computer Science and Engineering Nanjing 210096 China Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Application China Southeast University Jiangsu Key Laboratory of Biomaterials and Devices Nanjing 211189 China Southeast University State Key Laboratory of Digital Medical Engineering Nanjing 211189 China

One of the core challenges in ultrasound image quality assessment (IQA) is the entanglement of semantic content and quality-related information, such as blurring and shadows. Insufficient attention to the latter can easily lead to biased IQA results. Furthermore, fine-grained quality inconsistencies, i.e., subtle variations in ultrasound images that can impact quality interpretations, may further complicate the IQA tasks. To address these challenges, we propose a novel degradation-aware model (DAM) for the ultrasound IQA, which effectively perceives various and subtle variations of quality patterns, accurately assessing the quality of ultrasound images. The advanced degradation-derived augmentation (DDA) in DAM incorporates degradations that clinicians may focus on during IQA into the synthesis of appearance changes, promoting the disentanglement of quality-related representations from semantic contents. Subsequently, we present fine-grained degradation learning (FGDL), which encourages distinctions between image versions with diminishing quality inconsistencies, boosting the awareness of quality nuances from easy to hard for better ultrasound IQA performance. A universal boundary acquisition operator (UBAO) is also developed to suppress interferences from redundant information, achieving the standardization of ultrasound images from various devices. Extensive experimental results on an in-house ultrasound dataset demonstrate that DAM outperforms 14 baseline methods, achieving a PLCC of 0.760 and an SROCC of 0.766. © 2013 IEEE.

关键词： degradation-aware model degradation-derived augmentation fine-grained degradation learning Ultrasound image quality assessment universal boundary acquisition operator

来源：评论

学校读者我要写书评

暂无评论

Gaf-Net: Graph Attention Fusion Network for Multi-View Semi-Supervised Classification

SSRN

引用

SSRN 2023年

作者： Song, Na Du, Shide Wu, Zhihao Zhong, Luying Wang, Shiping School of Computer Science and Technology Hainan University Hainan570228 China School of Mechanical Electrical and Information Engineering Putian University Putian351100 China College of Computer and Data Science Fuzhou University Fuzhou350116 China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou350116 China

Multi-view semi-supervised classification is a typical task to classify data using a small amount of supervised information, which has attracted a lot of attention from researchers in past years. In practice, existing methods focus more on extracting spatial or spectral features using graph convolutional neural networks without considering the diversity and variability of graph structures and the contributions of different views. To address this challenge, a network framework termed graph attention fusion network is proposed, which is comprised of two phases: view-specific feature embedding and graph embedding fusion. In the former feature extraction stage, view-specific feature embedding module can flexibly focus on the neighborhood calculation operation to learn a weight for each neighboring node. In the latter feature fusion stage, the graph embedding fusion module is performed by complementarity and consistency to fuse these embeddings for semi-supervised classification tasks. We conduct extensive experiments in terms of semi-supervised classification on real-world datasets to validate the superiority of the proposed method over various existing state-ofthe-art methods. © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

EFFICIENT ONLINE LABEL CONSISTENT HASHING FOR LARGE-SCALE CROSS-MODAL RETRIEVAL

EFFICIENT ONLINE LABEL CONSISTENT HASHING FOR LARGE-SCALE CR...

引用

2021 IEEE International Conference on Multimedia and Expo, ICME 2021

作者： Yi, Jinhan Liu, Xin Cheung, Yiu-Ming Xu, Xing Fan, Wentao He, Yi Department of Computer Science and Technology Huaqiao University Xiamen361021 China Xiamen Key Lab. of Computer Vision and Pattern Recognition Fujian Key Lab. of Big Data Intelligence and Security China Department of Computer Science Hong Kong Baptist University Kowloon Hong Kong School of Computer Science and Engineering University of Electronic Science and Technology of China China Provincial Key Laboratory for Computer Information Processing Technology Soochow University China

ISBN: (纸本)9781665438643

Existing cross-modal hashing still faces three challenges: (1) Most batch-based methods are unsuitable for processing large-scale and streaming data. (2) Current online methods often suffer from insufficient semantic association, while lacking flexibility to learn the hash functions for varying streaming data. (3) Existing supervised methods always require much computation time or accumulate large quantization loss to learn hash codes. To address above challenges, we present an efficient Online Label Consistent Hashing (OLCH) for cross-modal retrieval, which aims to incrementally learn hash codes for the current arriving data, while updating the hash functions at a streaming manner. To be specific, an online semantic representation learning framework is designed to adaptively preserve the semantic similarity across different modalities, and a mini-batch online gradient descent approach associated with forward-backward splitting is developed to optimize the hash functions. Accordingly, the hash codes are adaptively learned online with the high discriminative capability, while avoiding high computation complexity to process the streaming data. Experimental results show its outstanding performance in comparison with the-state-of-arts. © 2021 IEEE computer Society. All rights reserved.

关键词： Hash functions

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：