检索结果-南通市图书馆

An attention-embedded GAN for SVBRDF recovery from a single image

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2023年第3期9卷 551-561页

作者： Zeqi Shi Xiangyu Lin Ying Song Zhejiang Sci-Tech University 2011 Collaborative Innovation Center for Garment Personal Customization of Zhejiang ProvinceHangzhou 310018China.

Learning-based approaches have made substantial progress in capturing spatially-varying bidirectional reflectance distribution functions(SVBRDFs)from a single image with unknown lighting and geometry.However,most existing networks only consider per-pixel losses which limit their capability to recover local features such as smooth glossy regions.A few generative adversarial networks use multiple discriminators for different parameter maps,increasing network complexity.We present a novel end-to-end generative adversarial network(GAN)to recover appearance from a single picture of a nearly-flat surface lit by flash.We use a single unified adversarial framework for each parameter map.An attention module guides the network to focus on details of the maps.Furthermore,the SVBRDF map loss is combined to prevent paying excess attention to specular highlights.We demonstrate and evaluate our method on both public datasets and real data.Quantitative analysis and visual comparisons indicate that our method achieves better results than the state-of-the-art in most cases.

关键词： spatially-varying bidirectional reflectance distribution function(SVBRDF) appearance capture generative adversarial network(GAN) attention mechanism

FlowDNN:a physics-informed deep neural network for fast and accurate flow prediction

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2022年第2期23卷 207-219页

作者： Donglin CHEN Xiang GAO Chuanfu XU Siqi WANG Shizhao CHEN Jianbin FANG Zheng WANG College of Computer National University of Defense TechnologyChangsha 410073China State Key Laboratory of High Performance Computing National University of Defense TechnologyChangsha 410073China School of Computing University of LeedsLeeds LS29JTUK

For flow-related design optimization problems,e.g.,aircraft and automobile aerodynamic design,computational fluid dynamics(CFD)simulations are commonly used to predict flow fields and analyze performance.While important,CFD simulations are a resource-demanding and time-consuming iterative process.The expensive simulation overhead limits the opportunities for large design space exploration and prevents interactive design.In this paper,we propose Flow DNN,a novel deep neural network(DNN)to efficiently learn flow representations from CFD results.Flow DNN saves computational time by directly predicting the expected flow fields based on given flow conditions and geometry shapes.Flow DNN is the first DNN that incorporates the underlying physical conservation laws of fluid dynamics with a carefully designed attention mechanism for steady flow prediction.This approach not only improves the prediction accuracy,but also preserves the physical consistency of the predicted flow fields,which is essential for CFD.Various metrics are derived to evaluate Flow DNN with respect to the whole flow fields or regions of interest(RoIs)(e.g.,boundary layers where flow quantities change rapidly).Experiments show that Flow DNN significantly outperforms alternative methods with faster inference and more accurate results.It speeds up a graphics processing unit(GPU)accelerated CFD solver by more than 14000×,while keeping the prediction error under 5%.

关键词： Deep neural network Flow prediction attention mechanism Physics-informed loss

DAAPS: A Deformable-attention-Based Anchor-Free Person Search Model

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2023年第11期77卷 2407-2425页

作者： Xiaoqi Xin Dezhi Han Mingming Cui School of Information Engineering Shanghai Maritime UniversityShanghai201306China

Person Search is a task involving pedestrian detection and person re-identification,aiming to retrieve person images matching a given objective attribute from a large-scale image library.The Person Search models need to understand and capture the detailed features and context information of smaller objects in the image more accurately and comprehensively.The current popular Person Search models,whether end-to-end or two-step,are based on anchor boxes.However,due to the limitations of the anchor itself,the model inevitably has some disadvantages,such as unbalance of positive and negative samples and redundant calculation,which will affect the performance of models.To address the problem of fine-grained understanding of target pedestrians in complex scenes and small sizes,this paper proposes a Deformable-attention-based Anchor-free Person Search model(DAAPS).Fully Convolutional One-Stage(FCOS),as a classic Anchor-free detector,is chosen as the model’s infrastructure.The DAAPS model is the first to combine the Anchor-free Person Search model with Deformable attention mechanism,applied to guide the model adaptively adjust the perceptual.The Deformable attention mechanism is used to help the model focus on the critical information and effectively improve the poor accuracy caused by the absence of anchor boxes.The experiment proves the adaptability of the attention mechanism to the Anchor-free model.Besides,with an improved ResNeXt+network frame,the DAAPS model selects the Triplet-based Online Instance Matching(TOIM)Loss function to achieve a more precise end-to-end Person Search task.Simulation experiments demonstrate that the proposed model has higher accuracy and better robustness than most Person Search models,reaching 95.0%of mean Average Precision(mAP)and 95.6%of Top-1 on the CUHK-SYSU dataset,48.6%of mAP and 84.7%of Top-1 on the Person Re-identification in the Wild(PRW)dataset,respectively.

关键词： Person Search anchor-free attention mechanism person detection pedestrian re-identification

Detecting broiler chickens on litter floor with the YOLOv5-CBAM deep learning model

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Artificial Intelligence in Agriculture

Artificial Intelligence in Agriculture 2023年第3期9卷 36-45页

作者： Yangyang Guo Samuel E.Aggrey Xiao Yang Adelumola Oladeinde Yongliang Qiao Lilong Chai School of Internet Anhui UniversityHefeiAnhui 230039China Department of Poultry Science University of GeorgiaAthensGA 30602USA U.S.National Poultry Research Center USDA-ARSAthensGA 30605USA Australian Centre for Field Robotics(ACFR) Faculty of EngineeringThe University of SydneyNSW 2006Australia

For commercial broiler production,about 20,000–30,000 birds are raised in each confined house,which has caused growing public concerns on animal welfare.Currently,daily evaluation of broiler wellbeing and growth is conducted manually,which is labor-intensive and subjectively subject to human error.Therefore,there is a need for an automatic tool to detect and analyze the behaviors of chickens and predict their welfare status.In this study,we developed a YOLOv5-CBAM-broiler model and tested its performance for detecting broilers on litter floor.The proposed model consisted of two parts:(1)basic YOLOv5 model for bird or broiler feature extraction and object detection;and(2)the convolutional block attention module(CBAM)to improve the feature extraction capability of the network and the problem of missed detection of occluded targets and small targets.A complex dataset of broiler chicken images at different ages,multiple pens and scenes(fresh litter versus reused litter)was constructed to evaluate the effectiveness of the new model.In addition,the model was compared to the Faster R-CNN,SSD,YOLOv3,EfficientDet and YOLOv5 models.The results demonstrate that the precision,recall,F1 score and an mAP@0.5 of the proposed method were 97.3%,92.3%,94.7%,and 96.5%,which were superior to the comparison models.In addition,comparing the detection effects in different scenes,the YOLOv5-CBAM model was still better than the comparison method.Overall,the proposed YOLOv5-CBAM-broiler model can achieve real-time accurate and fast target detection and provide technical support for the management and monitoring of birds in commercial broiler houses.

关键词： Poultry production Deep learning YOLOv5 attention mechanism

attention-Based CNN Fusion Model for Emotion Recognition During Walking Using Discrete Wavelet Transform on EEG and Inertial Signals

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2024年第1期7卷 188-204页

作者： Yan Zhao Ming Guo Xiangyong Chen Jianqiang Sun Jianlong Qiu School of Communication Engineering Hangzhou Dianzi UniversityHangzhou 310018China School of Automation and Electrical Engineering Linyi UniversityLinyi 276000China

Walking as a unique biometric tool conveys important information for emotion recognition.Individuals in different emotional states exhibit distinct walking patterns.For this purpose,this paper proposes a novel approach to recognizing emotion during walking using electroencephalogram(EEG)and inertial signals.Accurate recognition of emotion is achieved by training in an end-to-end deep learning fashion and taking into account multi-modal fusion.Subjects wear virtual reality head-mounted display(VR-HMD)equipment to immerse in strong emotions during walking.VR environment shows excellent imitation and experience ability,which plays an important role in awakening and changing emotions.In addition,the multi-modal signals acquired from EEG and inertial sensors are separately represented as virtual emotion images by discrete wavelet transform(DWT).These serve as input to the attention-based convolutional neural network(CNN)fusion model.The designed network structure is simple and lightweight while integrating the channel attention mechanism to extract and enhance features.To effectively improve the performance of the recognition system,the proposed decision fusion algorithm combines Critic method and majority voting strategy to determine the weight values that affect the final decision results.An investigation is made on the effect of diverse mother wavelet types and wavelet decomposition levels on model performance which indicates that the 2.2-order reverse biorthogonal(rbio2.2)wavelet with two-level decomposition has the best recognition performance.Comparative experiment results show that the proposed method outperforms other existing state-of-the-art works with an accuracy of 98.73%.

关键词： walking multi-modal fusion virtual reality emotion recognition discrete wavelet transform attention mechanism

Denoising Graph Inference Network for Document-Level Relation Extraction

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2023年第2期6卷 248-262页

作者： Hailin Wang Ke Qin Guiduo Duan Guangchun Luo School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China School of Computing and Artificial Intelligence Southwestern University of Finance and EconomicsChengdu 611130China

Relation Extraction(RE)is to obtain a predefined relation type of two entities mentioned in a piece of text,e.g.,a sentence-level or a document-level text.Most existing studies suffer from the noise in the text,and necessary pruning is of great importance.The conventional sentence-level RE task addresses this issue by a denoising method using the shortest dependency path to build a long-range semantic dependency between entity pairs.However,this kind of denoising method is scarce in document-level RE.In this work,we explicitly model a denoised document-level graph based on linguistic knowledge to capture various long-range semantic dependencies among entities.We first formalize a Syntactic Dependency Tree forest(SDT-forest)by introducing the syntax and discourse dependency relation.Then,the Steiner tree algorithm extracts a mention-level denoised graph,Steiner Graph(SG),removing linguistically irrelevant words from the SDT-forest.We then devise a slide residual attention to highlight word-level evidence on text and SG.Finally,the classification is established on the SG to infer the relations of entity pairs.We conduct extensive experiments on three public datasets.The results evidence that our method is beneficial to establish long-range semantic dependency and can improve the classification performance with longer texts.

关键词： Relation Eextraction(RE) document-level denoising linguistic knowledge attention mechanism

ARM3D:attention-based relation module for indoor 3D object detection

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2022年第3期8卷 395-414页

作者： Yuqing Lan Yao Duan Chenyi Liu Chenyang Zhu Yueshan Xiong Hui Huang Kai Xu College of Computer National University of Defense TechnologyChangsha 410073China Shenzhen University Shenzhen 518061China

Relation contexts have been proved to be useful for many challenging vision tasks.In the field of3D object detection,previous methods have been taking the advantage of context encoding,graph embedding,or explicit relation reasoning to extract relation contexts.However,there exist inevitably redundant relation contexts due to noisy or low-quality proposals.In fact,invalid relation contexts usually indicate underlying scene misunderstanding and ambiguity,which may,on the contrary,reduce the performance in complex scenes.Inspired by recent attention mechanism like Transformer,we propose a novel 3D attention-based relation module(ARM3D).It encompasses objectaware relation reasoning to extract pair-wise relation contexts among qualified proposals and an attention module to distribute attention weights towards different relation contexts.In this way,ARM3D can take full advantage of the useful relation contexts and filter those less relevant or even confusing contexts,which mitigates the ambiguity in detection.We have evaluated the effectiveness of ARM3D by plugging it into several state-of-the-art 3D object detectors and showing more accurate and robust detection results.Extensive experiments show the capability and generalization of ARM3D on 3D object detection.Our source code is available at https://***/lanlan96/ARM3D.

关键词： attention mechanism scene understanding relational reasoning 3D indoor object detection

Learning semantic-specific visual representation for laser welding penetration status recognition

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2022年第2期65卷 347-360页

作者： LIU TianYuan BAO JinSong ZHENG HangBin WANG JunLiang YANG ChangQi GU Jun Institute of Intelligent Manufacturing College of Mechanical EngineeringDonghua UniversityShanghai 201600China Shanghai Spaceflight Precision Machinery Institute Shanghai 201600China Shanghai Institute of Laser Technology Shanghai 200235China

The degree of penetration can directly reflect the forming quality of laser welding. The fine-grained feature of the molten pool/keyhole image brings challenges to the vision-based laser welding penetration status recognition. In this paper, a novel knowledge-and-data-hybrid driven recognition model is proposed for solving the problem of difficult learning of discriminative visual features of molten pool/keyhole images. In addition, a label semantic attention mechanism(LSA) is designed with three modules: representation of image visual feature, representation of labels semantic feature, and generation of label semantic attention. For learning discriminative features in visual space, LSA uses discriminative information in label semantics to guide the convolutional neural network. The experimental results show that the proposed LSA method has faster convergence and higher accuracy than the traditional attention mechanism. Further comparative experiments reveal that LSA is less dependent on the amount of training data and model complexity. The results of visualization experiments show that the visual features learned by the proposed method are more discriminative.

关键词： label semantic attention mechanism word embedding convolutional neural network laser welding penetration status

Material decomposition of spectral CT images via attention-based global convolutional generative adversarial network

在线全文

学校读者我要写书评

暂无评论

Nuclear Science and Techniques 2023年第3期34卷 143-153页

作者： Xiaodong Guo Peng He Xiaojie Lv Xuezhi Ren Yonghui Li Yuanfeng Liu Xiaohua Lei Peng Feng Hongming Shan The Key Lab of Optoelectronic Technology and Systems of the Education Ministry of China Chongqing UniversityChongqing 400044China Department of Vascular and Endovascular Surgery First Affiliated Hospital of Zhengzhou UniversityZhengzhou 450003China Institute of Science and Technology for Brain-inspired Intelligence and MOE Frontiers Center for Brain Science Fudan UniversityShanghai 200433China Shanghai Center for Brain Science and Brain-Inspired Technology Shanghai 200031China

Spectral computed tomography(CT)based on photon counting detectors can resolve the energy of every single photon interacting with the sensor layer and be used to analyze material attenuation information under different energy ranges,which can be helpful for material decomposition studies.However,there is a considerable amount of inherent quantum noise in narrow energy bins,resulting in a low signal-to-noise ratio,which can consequently affect the material decomposition performance in the image domain.Deep learning technology is currently widely used in medical image segmentation,denoising,and recognition.In order to improve the results of material decomposition,we propose an attention-based global convolutional generative adversarial network(AGC-GAN)to decompose different materials for spectral CT.Specifically,our network is a global convolutional neural network based on an attention mechanism that is combined with a generative adversarial network.The global convolutional network based on the attention mechanism is used as the generator,and a patchGAN discriminant network is used as the discriminator.Meanwhile,a clinical spectral CT image dataset is used to verify the feasibility of our proposed approach.Extensive experimental results demonstrate that AGC-GAN achieves a better material decomposition performance than vanilla U-Net,fully convolutional network,and fully convolutional denseNet.Remarkably,the mean intersection over union,structural similarity,mean precision,PAcc,and mean F1-score of our method reach up to 87.31%,94.83%,93.22%,97.39%,and 93.05%,respectively.

关键词： Photon-counting CT Material decomposition attention mechanism GAN