检索结果-内蒙古大学图书馆

Emerging image generation with flexible control of perceived difficulty

compUTER VISION AND IMAGE UNDERSTANDING 2024年 240卷

作者： Li, Jingmeng Wei, Hui Yang, Surun Fu, Lukang Fudan Univ Sch Comp Sci Lab Algorithms Cognit Models Shanghai Key Lab Data Sci 2005 Songhu Rd Shanghai 200438 Peoples R China

Emerging images (EI) are two-tone and contain a number of discrete speckles. If certain speckles are appropriately organized together, we will perceive a meaningful object, which reflects the closed -loop information processing of human visual cognition. EIs hold significant application value. They can be used in studies of perceptual organization in cognitive psychology. Additionally, they can also serve as a CAPTCHA mechanism to distinguish humans from bots in the field of network security. Both applications require a method for generating EIs that can flexibly adjust the perceived difficulty. Although universal style transfer (UST) models are capable of generating images in a specific style, it can be challenging to adjust the generated results to meet different user needs. In this paper, we present a novel EI generation framework and achieve flexible control over the perceived difficulty of the EIs by extracting and quantifying different cognitive cues and setting the corresponding parameters to adjust the proportion of these cues rendered in the EIs. The experimental results both qualitatively and quantitatively demonstrate that our methods generates EIs with higher quality while allowing for more flexible control over the perceived difficulty. Furthermore, we prove the potential of EIs as a CAPTCHA through sufficient experiments.

关键词： Visual emergence Closed-loop information processing Image Generation Emerging Image Style transfer CAPTCHAs

来源：评论

学校读者我要写书评

暂无评论

Multilevel network for large deformation image registration based on feature consistency and flow normalization

引用

MEDICAL PHYSICS 2024年第12期51卷 8962-8978页

作者： Huang, Xingyu Zhang, Jian Tang, Kun Cheng, Xinyu Ye, Chen Wang, Lihui Guizhou Univ Engn Res Ctr Text Comp & Cognit Intelligence Key Lab Intelligent Med Image Anal & Precise Diag Coll Comp Sci & TechnolMinist EducState Key Lab Guiyang Peoples R China

Background: Deformable image registration is an essential technique of medical image analysis, which plays important roles in several clinical applications. Existing deep learning-based registration methods have already achieved promising performance for the registrations with small deformations, while it is still challenging to deal with the large deformation registration due to the limits of the image intensity-similarity-based objective function. Purpose: To achieve the image registration with large-scale deformations, we proposed a multilevel network architecture FCNet to gradually refine the registration results based on semantic feature consistency constraint and flow normalization (FN) strategy. Methods: At each level of FCNet, the architecture is mainly composed to a FeaExtractor, a FN module, and a spatial transformation module. FeaExtractor consists of three parallel streams which are used to extract the individual features of fixed and moving images, as well as their joint features, respectively. Using these features, the initial deformation field is estimated, which passes through a FN module to refine the deformation field based on the difference map of deformation filed between two adjacent levels. This allows the FCNet to progressively improve the registration performance. Finally, a spatial transformation module is used to get the warped image based on the deformation field. Moreover, in addition to the image intensity-similarity-based objective function, a semantic-feature consistency constraint is also introduced, which can further promote the alignments by imposing the similarity between the fixed and warped image features. To validate the effectiveness of the proposed method, we compared our method with the state-of-the-art methods on three different datasets. In EMPIRE10 dataset, 20, 3, and 7 fixed and moving 3D computer tomography (CT) image pairs were used for training, validation, and testing respectively;in IXI dataset, atlas to individual i

关键词： coarse-to-fine deformable medical image registration feature consistency flow norm large deformations

来源：评论

学校读者我要写书评

暂无评论

High performance RGB-Thermal Video Object Detection via hybrid fusion with progressive interaction and temporal-modal difference

引用

INFORMATION FUSION 2025年 114卷

作者： Wang, Qishun Tu, Zhengzheng Li, Chenglong Tang, Jin Anhui Univ Sch Comp Sci & Technol Anhui Prov Key Lab Multimodal Cognit Computat Hefei 230601 Peoples R China Anhui Univ Sch Artificial Intelligence Anhui Prov Key Lab Multimodal Cognit Computat Hefei 230601 Peoples R China

RGB-Thermal Video Object Detection (RGBT VOD) is to localize and classify the predefined objects in visible and thermal spectrum videos. The key issue in RGBT VOD lies in integrating multi-modal information effectively to improve detection performance. Current multi-modal fusion methods predominantly employ middle fusion strategies, but the inherent modal difference directly influences the effect of multi-modal fusion. Although the early fusion strategy reduces the modality gap in the middle stage of the network, achieving indepth feature interaction between different modalities remains challenging. In this work, we propose a novel hybrid fusion network called PTMNet, which effectively combines the early fusion strategy with the progressive interaction and the middle fusion strategy with the temporal-modal difference, for high performance RGBT VOD. In particular, we take each modality as a master modality to achieve an early fusion with other modalities as auxiliary information by progressive interaction. Such a design not only alleviates the modality gap but facilitates middle fusion. The temporal-modal difference models temporal information through spatial offsets and utilizes feature erasure between modalities to motivate the network to focus on shared objects in both modalities. The hybrid fusion can achieve high detection accuracy only using three input frames, which makes our PTMNet achieve a high inference speed. Experimental results show that our approach achieves state-of-theart performance on the VT-VOD50 dataset and also operates at over 70 FPS. The code will be freely released at https://***/tzz-ahu for academic purposes.

关键词： Video object detection Multi-modal fusion RGB-thermal Temporal difference Hybrid strategy

来源：评论

学校读者我要写书评

暂无评论

A prototype evolution network for relation extraction

引用

APPLIED intelligence 2025年第1期55卷 1-14页

作者： Wang, Kai Chen, Yanping Huang, Ruizhang Qin, Yongbin Guizhou Univ Coll Comp Sci & Technol Text Comp & Cognit Intelligence Engn Res Ctr Natl Educ Minist Guiyang 550025 Guizhou Peoples R China Guizhou Univ State Key Lab Publ Big Data Guiyang 550025 Guizhou Peoples R China

Prototypical networks transform relation instances and relation types into the same semantic space, where a relation instance is assigned a type based on the nearest prototype. Traditional prototypical network methods generate relation prototypes by averaging the sentence representations from a predefined support set, which suffers from two key limitations. One limitation is sensitive to the outliers in the support set that can skew the relation prototypes. Another limitation is the lack of the necessary representational capacity to capture the full complexity of the relation extraction task. To address these limitations, we propose the Prototype Evolution Network (PEN) for relation extraction. First, we assign a type cue for each relation instance to mine the semantics of the relation type. Based on the type cues and relation instances, we then present a prototype refiner comprising a multichannel convolutional neural network and a scaling module to learn and refine the relation prototypes. Finally, we introduce historical prototypes during each episode into the current prototype learning process to enable continuous prototype evolution. We evaluate the PEN on the ACE 2005, SemEval 2010, and CoNLL2004 datasets, and the results demonstrate impressive improvements, with the PEN outperforming existing state-of-the-art methods.

关键词： Natural language processing Relation extraction Prototypical network Prototype evolution

来源：评论

学校读者我要写书评

暂无评论

Knowledge-driven multi-graph convolutional network for brain network analysis and potential biomarker discovery

引用

MEDICAL IMAGE ANALYSIS 2025年 99卷 103368页

作者： Zeng, Xianhua Gong, Jianhua Li, Weisheng Yang, Zhuoya Chongqing Univ Posts & Telecommun Sch Comp Sci & Technol Sch Artificial Intelligence Chongqing Key Lab Image Cognit Chongqing 400065 Peoples R China Chongqing Univ Posts & Telecommun Key Lab Cyberspace Big Data Intelligent Secur Minist Educ Chongqing 400065 Peoples R China

In brain network analysis, individual-level data can provide biological features of individuals, while population- level data can provide demographic information of populations. However, existing methods mostly utilize either individual- or population-level features separately, inevitably neglecting the multi-level characteristics of brain disorders. To address this issue, we propose an end-to-end multi-graph neural network model called KMGCN. This model simultaneously leverages individual- and population-level features for brain network analysis. At the individual level, we construct multi-graph using both knowledge-driven and data-driven approaches. Knowledge-driven refers to constructing a knowledge graph based on prior knowledge, while data-driven involves learning a data graph from the data itself. At the population level, we construct multi- graph using both imaging and phenotypic data. Additionally, we devise a pooling method tailored for brain networks, capable of selecting brain regions that impact brain disorders. We evaluate the performance of our model on two large datasets, ADNI and ABIDE, and experimental results demonstrate that it achieves stateof-the-art performance, with 86.87% classification accuracy for ADNI and 86.40% for ABIDE, accompanied by around 10% improvements in all evaluation metrics compared to the state-of-the-art models. Additionally, the biomarkers identified by our model align well with recent neuroscience research, indicating the effectiveness of our model in brain network analysis and potential biomarker discovery. The code is available at https: //***/GN-gjh/KMGCN.

关键词： Multi-graph convolution Knowledge graph Multi-level Biomarker

来源：评论

学校读者我要写书评

暂无评论

CEDNET: A CONTINUOUS EMOTION DETECTION NETWORK FOR NATURALISTIC STIMULI USING MEG SIGNALS 49

CEDNET: A CONTINUOUS EMOTION DETECTION NETWORK FOR NATURALIS...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： He, Zeming Zhang, Gaoyan Tianjin Univ Tianjin key Lab Cognit Comp & Applicat Coll Intelligence & Comp Tianjin Peoples R China

ISBN: (纸本)9798350344868;9798350344851

Emotional detection is important for brain-computer interface or diagnosis of affective disorders. Traditional methods mainly focused on the recognition of brief stimuli evoked emotion, which cannot fully represent the complexity of real-life emotional changes. In this study, we proposed a continuous emotion detection network (CEDNet) based on magnetoencephalography (MEG) to detect the time-varying emotions evoked by a 2-hour movie. Two brain graphs constructed by functional connectivity and spatial location of brain regions were input as two views of the brain. An adaptive spatio-temporal graph convolutional network with an attention mechanism was adopted to extract the emotion-related high-level features. Considering the impact of unbalanced emotion labels, a label distribution smoother was introduced. Furthermore, we added a domain discriminator to enhance the generalization capability of the model. Experimental results show that the proposed model outperforms the state-of-the-art baselines and provides a deep insight into the human emotional processes.

关键词： Continuous emotion detection naturalistic stimuli spatio-temporal graph convolution label distribution smoother MEG

来源：评论

学校读者我要写书评

暂无评论

Dialogue sentiment analysis based on dialogue structure pre-training

引用

MULTIMEDIA SYSTEMS 2025年第2期31卷 1-13页

作者： Yang, Liang Yang, Qi Zeng, Jingjie Peng, Tao Yang, Zhihao Lin, Hongfei Dalian Univ Technol Sch Comp Sci & Technol Dalian Peoples R China Dalian Univ Technol Minist Educ Key Lab Social Comp & Cognit Intelligence Dalian Peoples R China

The task of dialogue sentiment analysis aims to identify the sentiment polarity of utterances in the context of a dialogue. Pre-trained models often struggle to capture the logical structure of a dialogue, making this task challenging. To address this issue, we propose a dialogue sentiment analysis framework that leverages pre-training on dialogue structure. Our proposed framework includes three sub-tasks for pre-training: utterance order sorting, sentence backbone regularization, and sentiment shift detection. These tasks are designed to improve the model's ability to mine dialogue logical relationships and sentiment interactions. By focusing on learning the logical structure of dialogues and the perception of sentiment interactions, our framework is able to improve the performance of pre-trained models on recognizing the sentiment polarity of dialogues. This is demonstrated by the convincing results obtained on the public MEISD dataset.

关键词： Dialogue structure Sentiment analysis Pre-training model

来源：评论

学校读者我要写书评

暂无评论

Enhancing Major Depressive Disorder Diagnosis With Dynamic-Static Fusion Graph Neural Networks

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2024年第8期28卷 4701-4710页

作者： Zhao, Tianyi Zhang, Gaoyan Tianjin Univ Coll Intelligence & Comp Tianjin Key Lab Cognit Comp & Applicat Tianjin 300350 Peoples R China Univ Southern Calif Dept Comp Sci Los Angeles CA 90089 USA

Major Depressive Disorder (MDD) is a debilitating, complex mental condition with unclear mechanisms hindering diagnostic progress. Research links MDD to abnormal brain connectivity using functional magnetic resonance imaging (fMRI). Yet, existing fMRI-based MDD models suffer from limitations, including neglecting dynamic network traits, lacking interpretability, and struggling with small datasets. We present DSFGNN, a novel graph neural network framework addressing these issues for improved MDD diagnosis. DSFGNN employs a graph isomorphism encoder to model static and dynamic brain networks, achieving effective fusion of temporal and spatial information through a spatiotemporal attention mechanism, thereby enhancing interpretability. Furthermore, we incorporate a causal disentangling module and orthogonal regularization module to augment the model's expressiveness. We evaluate DSFGNN on the Rest-meta-MDD dataset, yielding superior results compared to the best baseline. Besides, extensive ablation studies and interpretability analysis confirm DSFGNN's effectiveness and potential for biomarker discovery.

关键词： Brain modeling Functional magnetic resonance imaging Depression Vectors Graph neural networks computational modeling Biological system modeling Dynamic brain functional connectivity functional magnetic resonance imaging graph neural networks major depressive disorder

来源：评论

学校读者我要写书评

暂无评论

SECaaS-Based Partially Observable Defense Model for IIoT Against Advanced Persistent Threats

引用

IEEE TRANSACTIONS ON SERVICES compUTING 2024年第6期17卷 4267-4280页

作者： Zhang, Zikai Ding, Chuntao Li, Yidong Yu, Jinhui Li, Jingyi Beijing Jiaotong Univ Key Lab Big Data & Artificial Intelligence Transpo Minist Educ Beijing 100044 Peoples R China Beijing Jiaotong Univ Sch Comp Sci & Technol Key Lab Big Data & Artificial Intelligence Transpo Minist Educ Beijing 100044 Peoples R China

With the advancement of intelligent and networked technology, the Industrial Internet of Things (IIoT) faces an escalating threat from cyberattacks, especially by Advanced Persistent Threat (APT) attacks. These novel and complex attacks, characterized by their dynamic nature and life-long duration, pose significant challenges to existing security protection methods. The challenges are twofold, i.e., sparse reward problem in the long-lasting attack, and partial observation of attack actions. To this end, we propose a Security-as-a-Service based reinforcement learning method, namely Attention Augmented Dueling Deep Q-learning Network (AD2QN), to make real-time defense strategies for the hot standby IIoT. First, we build the attack-defend confrontation model as black boxes interact with the IIoT environment to play a long-lasting partially observable zero-sum stochastic game on the server. Then, to dynamically generate optimal defense strategies as the service, AD2QN is proposed employing information completion and prediction to more informed action selection. Furthermore, AD2QN utilizes an iteratively updated reward network to deal with the sparse reward problem. Extensive simulation results shown that the defense strategies generated by our method have a higher defense success rate and a stable defense performance with the average success rate of 0.7384, while the average success rate of baseline methods was 0.7375, in the best case.

关键词： Industrial Internet of Things Security Games Stochastic processes Q-learning computer crime Uncertainty Advanced persistent threat attack-defend confrontation decision-making partially observable reinforcement learning security-as-a-service sparse reward

来源：评论

学校读者我要写书评

暂无评论

Geometry-semantic aware for monocular 3D Semantic Scene completion

引用

PATTERN REcognitION 2025年 158卷

作者： Lu, Zonghao Cao, Bing Xia, Shuyin Hu, Qinghua Tianjin Univ Coll Intelligence & Comp Tianjin 300350 Peoples R China Chongqing Univ Posts & Telecommun Key Lab Big Data Intelligent Comp Chongqing 400065 Peoples R China

Monocular Semantic Scene completion (SSC) empowers intelligent devices to comprehend voxel occupancy (geometry) and semantics in 3D scenes, attracting significant attention in indoor and autonomous driving scenarios. However, existing monocular SSC models primarily map 2D images into 3D space, neglecting the potential benefits of leveraging semantic and geometric understanding in 2D. To address this, we propose the Proxy-embedding Parallel Multi-task Network (PPMNet), which aims to perceive the geometry and semantics of 3D space through depth estimation and semantic segmentation proxy tasks on the 2D perspective plane. Moreover, 2D plane features can be inversely projected into 3D space and subsequently processed using the 3D network. In addition, we enhance contextual awareness in both perspective planes and voxel grids through parallel 2D and 3D decoders. Furthermore, we employ Dual-Head Pyramid Pooling (DHPP) to aggregate information from these two representations. Finally, considering the class imbalance and label incompleteness in practical data, we design a local-to-global loss to prioritize challenging categories. Extensive experiments validate our superiority over state-of-the-art methods on the NYUv2 and SemanticKITTI datasets. The code is available at: https://***/luzonghao1/PPMNet.

关键词： Semantic scene completion Depth estimation Semantic segmentation Contextual information

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：