检索结果-南通市图书馆

DepthGAN: GAN-based depth generation from semantic layouts

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第3期10卷 505-522页

作者： Yidi Li jun xiao Yiqun Wang Zhengda lu School of Artificial Intelligence University of Chinese Academy of SciencesBeijingChina College of Computer Science Chongqing UniversityChongqingChina

Existing GAN-based generative methodsare typically used for semantic image synthesis. Wepose the question of whether GAN-based architecturescan generate plausible depth maps and find thatexisting methods have difficulty in generating depthmaps which reasonably represent 3D scene structuredue to the lack of global geometric ***, we propose DepthGAN, a novel method ofgenerating a depth map using a semantic layout asinput to aid construction, and manipulation of wellstructured 3D scene point clouds. Specifically, wefirst build a feature generation model with a cascadeof semantically-aware transformer blocks to obtaindepth features with global structural *** our semantically aware transformer block, wepropose a mixed attention module and a semanticallyaware layer normalization module to better exploitsemantic consistency for depth features ***, we present a novel semantically weighteddepth synthesis module, which generates adaptivedepth intervals for the current scene. We generate thefinal depth map by using a weighted combination ofsemantically aware depth weights for different depthranges. In this manner, we obtain a more accuratedepth map. Extensive experiments on indoor andoutdoor datasets demonstrate that DepthGAN achievessuperior results both quantitatively and visually for thedepth generation task.

关键词： depth map generation generative model transformer scene generation

Feature Matching via Topology-Aware Graph Interaction Model

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 113-130页

作者： Yifan lu Jiayi Ma xiaoguang Mei jun Huang xiao-Ping Zhang the Electronic Information School Wuhan UniversityWuhan 430072China IEEE Tsinghua Berkeley Shenzhen Institute Shenzhen 518071China the Department of Electrical Computer and Biomedical EngineeringToronto Metropolitan UniversityToronto ON M5B 2K3Canada

Feature matching plays a key role in computer vision. However, due to the limitations of the descriptors, the putative matches are inevitably contaminated by massive *** paper attempts to tackle the outlier filtering problem from two aspects. First, a robust and efficient graph interaction model,is proposed, with the assumption that matches are correlated with each other rather than independently distributed. To this end, we construct a graph based on the local relationships of matches and formulate the outlier filtering task as a binary labeling energy minimization problem, where the pairwise term encodes the interaction between matches. We further show that this formulation can be solved globally by graph cut algorithm. Our new formulation always improves the performance of previous localitybased method without noticeable deterioration in processing time,adding a few milliseconds. Second, to construct a better graph structure, a robust and geometrically meaningful topology-aware relationship is developed to capture the topology relationship between matches. The two components in sum lead to topology interaction matching(TIM), an effective and efficient method for outlier filtering. Extensive experiments on several large and diverse datasets for multiple vision tasks including general feature matching, as well as relative pose estimation, homography and fundamental matrix estimation, loop-closure detection, and multi-modal image matching, demonstrate that our TIM is more competitive than current state-of-the-art methods, in terms of generality, efficiency, and effectiveness. The source code is publicly available at http://***/Yifanlu2000/TIM.

关键词： Feature matching graph cut outlier filtering topology preserving

CNN-LSTM based incremental attention mechanism enabled phase-space reconstruction for chaotic time series prediction

在线全文

学校读者我要写书评

暂无评论

Journal of Electronic Science and Technology 2024年第2期22卷 77-90页

作者： xiao-Qian lu jun Tian Qiang Liao Zheng-Wu Xu lu Gan Bidding and Purchasing Center University of Electronic Science and Technology of ChinaChengdu611731China Yibin Research Institute University of Electronic Science and Technology of ChinaYibin644000China School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengdu611731China

To improve the prediction accuracy of chaotic time series and reconstruct a more reasonable phase space structure of the prediction network,we propose a convolutional neural network-long short-term memory(CNN-LSTM)prediction model based on the incremental attention ***,a traversal search is conducted through the traversal layer for finite parameters in the phase ***,an incremental attention layer is utilized for parameter judgment based on the dimension weight criteria(DWC).The phase space parameters that best meet DWC are selected and fed into the input ***,the constructed CNN-LSTM network extracts spatio-temporal features and provides the final prediction *** model is verified using Logistic,Lorenz,and sunspot chaotic time series,and the performance is compared from the two dimensions of prediction accuracy and network phase space ***,the CNN-LSTM network based on incremental attention is compared with long short-term memory(LSTM),convolutional neural network(CNN),recurrent neural network(RNN),and support vector regression(SVR)for prediction *** experiment results indicate that the proposed composite network model possesses enhanced capability in extracting temporal features and achieves higher prediction ***,the algorithm to estimate the phase space parameter is compared with the traditional CAO,false nearest neighbor,and C-C,three typical methods for determining the chaotic phase space *** experiments reveal that the phase space parameter estimation algorithm based on the incremental attention mechanism is superior in prediction accuracy compared with the traditional phase space reconstruction method in five networks,including CNN-LSTM,LSTM,CNN,RNN,and SVR.

关键词： Chaotic time series Incremental attention mechanism Phase-space reconstruction

Experimental study of the damage characteristics of rocks containing non-penetrating cracks under cyclic loading

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

International Journal of Mining Science and Technology 2024年第2期34卷 197-210页

作者： jun Xu xiaochun xiao lu Ma Sen luo Jiaxu Jin Baijian Wu School of Mechanics and Engineering Liaoning Technical UniversityFuxin 123000China School of Civil Engineering Liaoning Technical UniversityFuxin 123000China Department of Engineering Mechanics Southeast UniversityNanjing 211189China

The damage evolution process of non-penetrating cracks often causes some unexpected engineering *** specimens containing non-penetrating crack(s)are used to study the damage evolution and characteristics under cyclic *** results show that under cyclic loading,the relationship between the number of non-penetrating crack(s)and the characteristic parameters(cyclic number,peak stress,peak strain,failure stress,and failure strain)of the pre-cracked specimens can be represented by a decreasing linear *** damage evolution equation is fitted by calibrating the accumulative plastic strain for each cycle,and the damage constitutive equation is proposed by the concept of effective ***,non-penetrating cracks are more likely to cause uneven stress distribution,damage accumulation,and local failure of *** local failure can change the stress distribution and relieve the inhibition of non-penetrating crack extension and eventually cause a dramatic destruction of the ***,the evolution process caused by non-penetrating cracks can be regarded as one of the important reasons for inducing *** results are expected to improve the understanding of the process of spalling formation and rockburst and can be used to analyze the stability of rocks or rock structures.

关键词： Damage characteristics Constitutive model Fissured rocks Non-penetrating crack Cyclic loading

Bang Bang Control of Unified Chaotic System

在线全文

学校读者我要写书评

暂无评论

Wuhan University Journal of Natural Sciences 2003年第S1期8卷 311-315页

作者： Deng xiao-ming, lu jun-anSchool of Mathematics and Statistics,Wuhan University, Wuhan 430072,Hubei,China School of Mathematics and Statistics Wuhan University Wuhan Hubei China

The unified chaotic system contains the Lorenz system and the Chen system as two dual systems at the two extremes of its parameter spectrum. This paper presents the design of bang bang controller for unified system and multitude of numerical experiments under various control parameters. Numerical experiments meet the theoretic proof perfectly and convincingly demonstrated the controller can be effectively used for unified systems with uncertainty of the equilibrium points. The method enriches the applications of chaotic control.

关键词： chaotic control bang bang control unified system

Synchronization of Unified Chaotic System Using Occasional Driving

在线全文

学校读者我要写书评

暂无评论

Wuhan University Journal of Natural Sciences 2003年第3A期8卷 808-812页

作者： Wu xiao\|qun, lu jun\|an  School of Mathematics and Statistics, Wuhan University , Wuhan 430072, Hubei, China School of Mathematics and Statistics Wuhan University Wuhan Hubei China

The synchronization of the unified chaotic systems using occasional driving technique is studied. The relation among interim period T , sampling interval 2 ε , feedback gain r and the parameter α of the system is thoroughly investigated. Numerical results show that smaller interim period T and properly larger sampling interval 2 ε can accelerate the synchronizing pace. Furthermore, for a unified chaotic system in which α is given, we can achieve satisfying synchronizing results as long as T,ε and r are appropriately chosen. As we adopt the occasional driving method, we greatly reduce the control cost. Therefore with this method we can obtain the expecting goals with little control cost.

关键词： synchronization occasional drive unified chaotic system

Laparoscopic adrenocortical oncocytoma resection

在线全文

学校读者我要写书评

暂无评论

Chinese Medical Journal 2008年第9期121卷 862-864页

作者： TIAN xiao-jun HUANG Yi lu Jian MA lu-lin Depatment of Urology Peking University Third Hospital Beijing 100083 China

The term ＂oncocyte＂ was first used by Hamperl in 1950 to describe a kind of large, highly eosinophilic, granular tumour cell. Oncocytoma refers to the tumour that is composed of oncocytes. Oncocytoma can occur in various organs throughout the body, commonly salivary glands, thyroid or kidney and less commonly in lung, parathyroid gland, ovary or pituitary gland. Adrenocortical oncocytomas are rare： only 46 cases have been reported worldwide in the literature.2-6 Most of adrenal masses were resected by open operations.3-7 An oncocytoma of the adrenal gland was diagnosed and treated in our hospital recently. Retroperitoneal laparoscopy was performed to remove the mass. We report our recent case and review the literature.

关键词： adrenal glands adenoma oxyphilic laparoscopy

Pre-peritoneal laparoscopic partial cystectomy of the bladder pheochromocytoma

在线全文

学校读者我要写书评

暂无评论

Chinese Medical Journal 2009年第10期122卷 1234-1237页

作者： HUANG Yi TIAN xiao-jun MA lu-lin Department of Urology Peking University Third Hospital Beijing 100191 China

Pheochromocytoma （PCC） is a catecholamine producing tumor of neuroectodermal origin that is made up of chromaftin cells, anti it mostly arises from me adrenal medulla. Extra-adrenal pheochromocytomas occur in about ... 详细信息

关键词： pheochromocytoma laparoscopy partial cystectomy

Review on regulation of MnS in non-qquenched and tempered steel

在线全文

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2024年第4期31卷 779-789页

作者： Guo-xing Qiu Qing Du Feng lu De-jun Miao Yong-kun Yang xiao-ming Li School of Metallurgical Engineering Xi'an University of Architecture and TechnologyXi'an 710055ShaanxiChina Laigang Technique Center Shandong Iron and Steel Group Co.Ltd.Jinan 271104ShandongChina

An overview of the current research status and control methods of MnS in non-quenched and tempered steel was *** a low-melting plastic inclusion,the morphology and distribution of MnS were influenced by various production ***,control of MnS is a systematic problem that must be integrated into the entire production *** on the production process,the factors affecting the morphology and distribution of MnS in steel were *** effects of oxygen activity,manganese,sulfur,and some alloys on MnS inclusion precipitation were summarized,mainly including MnS modification treatment and oxygen-sulfide composite precipitation *** is believed that MnS precipitates during the solidification process of steel,and controlling the solidification cooling rate could effectively regulate the size and morphology of MnS,avoiding the precipitation of ***,by changing the deformation rate,deformation amount,deformation temperature during the hot deformation process,and heating time and temperature during heat treatment,the distribution and morphology of MnS could be *** the fine control of the above process parameters,the number of II-MnS in steel could be effectively reduced,and their morphology could be improved,thereby enhancing the performance of non-quenched and tempered steel and promoting its wider ***,applying laboratory research results to industrial production is an important direction for future research efforts in this field.

关键词： MnS Non-quenched and tempered steel Control technology Process parameter Morphology Distribution