检索结果-内蒙古大学图书馆

SeaCap: Multi-Sight Embedding and Alignment for One-Stage Image Captioner

IEEE Transactions on Multimedia 2025年 27卷 3411-3425页

作者： Wang, Bo Zhang, Zhao Zhao, Mingbo Jin, Xiaojie Xu, Mingliang Wang, Meng Hefei University of Technology School of Computer Science and Information Engineering Hefei230601 China Yunnan Key Laboratory of Software Engineering Yunan China Donghua University School of Computer Science and Engineering Shanghai201620 China ByteDance Culver CityCA90230 United States Zhengzhou University School of Computer and Artificial Intelligence Zhengzhou450001 China

Recent mainstream image captioning methods usually adopt two-stage captioners, i.e., calculating the object features of the given image by a pre-trained detector and then feeding them into a language model to generate the descriptive sentences. However, such a two-stage procedure will lead to a task-based information gap that decreases the performance of the captioners, because the object features learned from the detection task are suboptimal representations and cannot provide all the necessary information for subsequent sentence generation. Besides, the object features are usually represented by the last pooling features of the detector that lose the local details of images. In this paper, we propose a novel One-Stage Image Captioner using dynamic multi-sight embedding and alignment, called SeaCap, which directly transforms input images into descriptive sentences in one stage to eliminate the information gap. Specifically, to obtain rich features, we use the Swin Transformer to capture multi-level features, followed by a sights alignment module to alleviate the vision confusion, and then feed them into a novel dynamic multi-sight embedding module to exploit both the global structure and local texture of input images. To enhance the global modeling capacity of the visual encoder, we propose a new dual-dimensional refining module to non-locally model the interaction of the embedded features. As a result, SeaCap can obtain rich and useful information to improve the performance of the captioner. Extensive comparisons on the benchmark MS-COCO, Flickr8K and Flickr30 K datasets verified the superior performance of our method. © 1999-2012 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Research on Online Education Model and Crisis Intervention System in Universities Perspective on Mental Health 22

Research on Online Education Model and Crisis Intervention S...

引用

6th International Conference on Education and Multimedia technology, ICEMT 2022

作者： Tai, Ruiting Chen, Chizhi Hwang, Zhiheng School of Information and Software Engineering University of Electronic Science and Technology of China China

ISBN: (纸本)9781450396455

Due to COVID-19, online education has become the main and new teaching method in the lockdown zone. However, teachers and students inevitably feel difficult and distressful about these new ways. There are also some other issues, such as inadequate interactions in the online classroom, causing poor knowledge absorption. Compared with traditional offline education, students cannot release their emotions or psychological pressure. What's more, teachers aren't able to comprehensively evaluate students' psychological conditions, which lead to belated awareness and solutions of their psychological problems. Therefore, from the perspective of mental health, this paper is designed to analyze the needs of college students for online education based on Maslow's hierarchy of needs theory, and to explore the corresponding mode and interventional strategies of psychological crisis under the impact of COVID-19 epidemic. Equipped with emotion recognition and attention recognition based on computer vision technology, an intelligent interactive online classroom is constructed. At the same time, by using social big data, natural language processing (NLP) emotion analysis is achieved, which has made multi-index mental health prediction and crisis intervention realized. Although under the influence of COVID-19, this will still help educators implement more comprehensive and convenient online education and provides a guarantee for the overall development for students. © 2022 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Automatic Generation of Formal BIP Models for C Programs 23

Automatic Generation of Formal BIP Models for C Programs

引用

7th International Conference on Computer Science and Application engineering, CSAE 2023

作者： Li, Qiming Yang, Xia Sun, Haiyong Yan, Zhe School of Information and Software Engineering University of Electronic Science and Technology of China Sichuan China Chengdu Lianan Tech Sichuan China

ISBN: (纸本)9798400700590

The BIP model is based on a combinatorial model-checking technique, which decomposes a software system into behavioral components, interaction components and priority rules. It can effectively formalize the description and verification of complex software systems, and is widely used in industry. However, BIP models require manual modeling and rely heavily on professional knowledge, so there are issues related to high cost, low efficiency, and difficulty in guaranteeing model accuracy. In this paper, an automatic BIP model generation method for C programs is proposed, which can automatically extract key program information to build a middle layer Labelled Transition System (LTS) and finally generate a BIP model. This paper further conducts automatic behavioral modeling of a high-performance dynamic memory allocation algorithm and successfully verifies 12 essential security requirements related to the algorithm. Experimental results show that the proposed automatic formal modeling method can effectively reduce manual dependency and improve modeling efficiency, while ensuring the accuracy of the constructed model. © 2023 ACM.

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

Dynamic View Synthesis of Thin Structures with Short-term Movements from Monocular Videos Using Neural Radiance Fields 25

Dynamic View Synthesis of Thin Structures with Short-term Mo...

引用

25th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2024

作者： Rajapaksha, Uchitha Laga, Hamid Diepeveen, Dean Bennamoun, Mohammed Sohel, Ferdous School of Information Technology Murdoch University MurdochWA Australia The University of Western Australia Dept. of Computer Science and Software Engineering CrawleyWA Australia

ISBN: (纸本)9798350379037

Learning to generate motions of thin structures such as plant leaves in dynamic view synthesis is challenging. This is because thin structures usually undergo small but fast, non-rigid motions as they interact with air and wind. When given a set of RGB images or videos of a scene with moving thin structures as input, existing methods that map the scene to its corresponding canonical space for rendering novel views fail as the object movements are too subtle compared to the background. Disentangling the objects with thin parts from the background scene is also challenging when the parts show fast and rapid motions. To address these issues, we propose a Neural Radiance Field (NeRF)-based framework that accurately reconstructs thin structures such as leaves and captures their subtle, fast motions. The framework learns the geometry of a scene by mapping the dynamic images to a canonical scene in which the scene remains static. We propose a ray masking network to further decompose the canonical scene into foreground and background, thus enabling the network to focus more on foreground movements. We conducted experiments using a dataset containing thin structures such as leaves and petals, which include image sequences collected by us and one public image sequence. Experiments show superior results compared to existing methods. Video outputs are available at https://***/. © 2024 IEEE.

关键词： Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

Strong Convective Weather Extrapolation Based on VAN-ConvLSTM 7

Strong Convective Weather Extrapolation Based on VAN-ConvLST...

引用

2022 7th International Conference on Electronic technology and information Science, ICETIS 2022

作者： Wei, Xiaofei Wen, Liyu Luo, Fei Department of Software Engineering Chengdu University of Information Technology Sichuan Chengdu China

ISBN: (纸本)9783800757794

Strong convective weather forecasting is an extension of traditional weather forecasting. The accuracy of short-term forecasts of strong convective weather becomes particularly important in early warnings. However, existing forecasts of severe convective weather often have difficulty responding accurately and quickly to weather hazards, while predictions are relatively inaccurate. In this paper, we propose a deep learning model. The model mixes multilayer architectures, including a deep visual analog making network (VAN) and a convolutional long and short term memory network (ConvLSTM). We use ConvLSTM to process the spatial features of the temporal images, while adding VAN to perform image analogy on the single frame image passed into the network. With this method, a continuous image of the same spatial output can be obtained based on a time series. The experimental results show that the prediction model combining VAN and ConvLSTM proposed in this paper has better prediction performance than other existing methods. © VDE VERLAG GMBH ∙ Berlin ∙ Offenbach.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Dysfunctional brain dynamics in Subjects with major depression: An EEG microstate spectral analysis

Dysfunctional brain dynamics in Subjects with major depressi...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Li, Jianxiu Hao, Yanrong Xu, Hongwei Inner Mongolia University School of Electronic Information Engineering Hohhot China Taiyuan University of Technology School of Software Taiyuan China

ISBN: (纸本)9798350337488

Major depressive disorder (MDD) which is a widespread disorder worldwide cause disruption in some brain functions and thus leads to brain network changes. An increasing number of clinical and cognitive neuroscience studies have used broadband EEG microstate method to assess the electrical activity of large-scale cortical networks;however, the topographic frequency patterns of the different EEG microstate categories in MDD patients are not clear. In this study, EEG microstate frequency spectra were analyzed using 5-min resting-state electroencephalography (EEG) data from 55 subjects (including 27 MDD patients and 28 healthy controls) with variational empirical mode decomposition in Hilbert-Huang transform. The results showed that microstate D and the other microstates (A, C, and E) display opposite patterns. Most specifically, in the beta band, the marginal spectral energies of microstates A, C, and E were increased in MDD patients compared to HCs, while the energy of microstate D was decreased. Meanwhile, we observed that the marginal spectral energies of microstates A and C in the beta band were positively correlated with the severity of depressive symptoms, suggesting that alterations in the beta band energy can serve as an important reference for disease progression. These results confirm the abnormalities of beta-band energy in MDD patient and provide a new perspective for exploring the abnormalities of psychomotor function in patients with MDD. © 2023 IEEE.

关键词： Brain network dynamics Electroencephalography MDD Microstate spectral analysis

来源：评论

学校读者我要写书评

暂无评论

Motion path planning method based on map reference points

引用

Multimedia Tools and Applications 2024年 1-21页

作者： Yu, Wenbin Fan, Cheng Zhang, Chengjun Chen, Yadang Li, Yangsong Zhang, Yifan Yin, Na Cen, Xiaolin School of Software Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China School of Computer Science Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China Wuxi Institute of Technology Nanjing University of Information Science and Technology Jiangsu Wuxi214000 China Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China School of Computer Science Northeastern University Boston02101 United States Computer Science and Engineering The University of New South Wales SydneyNSW02101 Australia

In recent years, significant progress has been made in the development of the Rapidly-exploring Random Trees (RRT) algorithm, a probabilistic sampling-based search approach. This algorithm effectively addresses complex challenges in path planning within high-dimensional spaces characterized by intricate constraints. However, the RRT algorithm faces certain limitations, including sluggish computational performance in handling intricate calculations, susceptibility to becoming trapped in confined path regions, and difficulties in generating randomized paths across a given map. To overcome these issues, this study introduces a motion path planning method based on map reference points. Initially, the algorithm selects strategic map reference points, followed by the utilization of an enhanced RRT algorithm for fundamental path selection. Real-world scenario data is collected to empirically evaluate the algorithm’s performance. The outcomes of the experiments reveal that, within the same experimental setting, the proposed method outperforms conventional motion path planning algorithms by effectively circumventing obstacles. Furthermore, it substantially reduces the resource wastage caused by global random sampling in the RRT algorithm, leading to expedited path planning and improved pathfinding success rates. Additionally, the proposed approach enhances the RRT algorithm’s adaptability to some extent, particularly in negotiating narrow passages and intricate environments. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Improved AODNet for Fast Image Dehazing 13th

Improved AODNet for Fast Image Dehazing

引用

13th EAI International Conference on Mobile Networks and Management, MONAMI 2023

作者： Chen, Shiyu Liu, Shumin Chen, Xingfeng Dan, Jiannan Wu, Bingbing School of Software Engineering Jiangxi University of Science and Technology Nanchang330013 China Aerospace Information Research Institute Chinese Academy of Sciences Beijing100101 China

ISBN: (纸本)9783031554704

Application scenarios such as unmanned driving and UAV reconnaissance have the requirements of high performance, low delay and small space occupation. Images taken in foggy days are easy to be affected by fog or haze, thus losing some important information. The purpose of image dehazing is to remove the influence of fog on image quality, which is of great significance to assist in solving high-level vision tasks. Aiming at the shortcomings of the current defogging method, such as slow defogging speed and poor defogging effect, this paper introduces the idea of FPCNet and the attention mechanism module, and proposes an improved AODNet fast defogging algorithm to ensure the defogging speed and defogging performance. The public dataset RESIDE was used for training and testing. Experimental results show that in terms of dehazing performance, the proposed algorithm achieves 25.78 and 0.992 in PSNR and SSIM respectively. In terms of dehazing speed, the proposed method is close to AODNet, with only 5 times more parameters than AODNet, but more than 100 times smaller than other methods. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

ASIMSA: Advanced Semantic information Guided Multi-Scale Alignment Framework for Medical Vision-Language Pretraining 9

ASIMSA: Advanced Semantic Information Guided Multi-Scale Ali...

引用

9th IEEE International Conference on Computational Intelligence and Applications, ICCIA 2024

作者： Xiao, Shuai Zhang, Yangyang Jiang, Liming Wang, Zhengxia School of Computer Science and Technology Hainan University China School of Information and Communication Engineering Hainan University China Software Co. Ltd. China

ISBN: (纸本)9798350352214

Medical Visual Language Pretraining (MVLP) utilizes textual reports for weak supervision to improve the learning of medical visual representations, showing promise in various medical image analysis tasks. While previous studies have focused on image and report correspondences within the same instances, they have overlooked high-level semantic associations, such as disease-level semantic correspondences, across different instances. To address this issue, we propose an Advanced Semantic information Guided Multi-Scale Alignment Framework (ASIMSA) with two main modules: 1) Image-Report Alignment (IRA) module for learning shared representations between image and report of the same instance, and 2) Disease-Entity Alignment (DEA) module for learning disease entity correspondences across different instances. Extensive experiments on zero-shot classification, fine-tune classification, and semantic segmentation tasks validate the effectiveness of our proposed method, demonstrating its stable and superior performance. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Investigation of Optimisation of PID-Controlled Electrochromic Device Voltage Using Genetic Algorithm 11

Investigation of Optimisation of PID-Controlled Electrochrom...

引用

11th International Forum on Electrical engineering and Automation, IFEEA 2024

作者： Yao, Qingxin Wu, Yitong Chen, Ziyu Kong, Sifan Wu, Hanmao Yang, Huan School of Electronics and Information Engineering South China Normal University Foshan528225 China School of Electronics and Information Engineering Harbin Institute of Technology Shenzhen518055 China School of Software South China Normal University Foshan528225 China

ISBN: (纸本)9798331516611

Electrochromic devices (ECDs) hold significant potential for applications in smart windows and automotive rearview mirrors. However, during continuous charge-discharge cycles, these devices are prone to charge accumulation, leading to increased potential differences. To effectively address this technical challenge, we propose a PID control strategy optimized by a genetic algorithm (GA). This strategy precisely regulates the voltage, ensuring that ECDs maintain a stable voltage level of target voltage during operation, thereby enhancing device stability and extending lifespan. This study innovatively applies the PID algorithm to the voltage control of ECDs and utilizes GA for adaptive PID parameter adjustment to improve control accuracy and response speed. Experimental results demonstrate that ECDs utilizing the optimized PID control strategy exhibit exceptional voltage stability, providing crucial support for the further advancement and application of electrochromic technology. Additionally, the adaptive PID controller model proposed in this work is developed using MATLAB/SIMULINK, offering new research methods and practical approaches for voltage control in ECDs. © 2024 IEEE.

关键词： MATLAB

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：