检索结果-南通市图书馆

One-shot Face Reenactment with Dense Correspondence Estimation

在线全文

学校读者我要写书评

暂无评论

Machine Intelligence research 2024年第5期21卷 941-953页

作者： Yunfan Liu Qi Li Zhenan Sun School of Artificial Intelligence University of Chinese Academy of SciencesBeijing100049China Center for research on Intelligent Perception and Computing Institute of AutomationChinese Academy of SciencesBeijing100190China

One-shot face reenactment is a challenging task due to the identity mismatch between source and driving *** existing methods fail to completely eliminate the interference of driving subjects’identity information,which may lead to face shape distortion and undermine the realism of reenactment *** solve this problem,in this paper,we propose using a 3D morphable model(3DMM)for explicit facial semantic decomposition and identity *** of using 3D coefficients alone for reenactment control,we take advantage of the generative ability of 3DMM to render textured face *** proxies contain abundant yet compact geometric and semantic information of human faces,which enables us to compute the face motion field between source and driving images by estimating the dense *** this way,we can approximate reenactment results by warping source images according to the motion field,and a generative adversarial network(GAN)is adopted to further improve the visual quality of warping *** experiments on various datasets demonstrate the advantages of the proposed method over existing state-of-the-art benchmarks in both identity preservation and reenactment fulfillment.

关键词： Generative adversarial networks face image manipulation face image synthesis face reenactment 3D morphable model

A Knowledge-enhanced Two-stage Generative Framework for Medical Dialogue Information 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

Machine Intelligence research 2024年第1期21卷 153-168页

作者： Zefa Hu Ziyi Ni Jing Shi Shuang Xu Bo Xu School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China institute of automation Chinese Academy of SciencesBeijing 100190China

This paper focuses on term-status pair extraction from medical dialogues(MD-TSPE),which is essential in diagnosis dia-logue systems and the automatic scribe of electronic medical records(EMRs).In the past few years,works on MD-TSPE have attracted increasing research attention,especially after the remarkable progress made by generative ***,these generative methods output a whole sequence consisting of term-status pairs in one stage and ignore integrating prior knowledge,which demands a deeper un-derstanding to model the relationship between terms and infer the status of each *** paper presents a knowledge-enhanced two-stage generative framework(KTGF)to address the above *** task-specific prompts,we employ a single model to com-plete the MD-TSPE through two phases in a unified generative form:We generate all terms the first and then generate the status of each generated *** this way,the relationship between terms can be learned more effectively from the sequence containing only terms in the first phase,and our designed knowledge-enhanced prompt in the second phase can leverage the category and status candidates of the generated term for status ***,our proposed special status"not mentioned"makes more terms available and en-riches the training data in the second phase,which is critical in the low-resource *** experiments on the Chunyu and CMDD datasets show that the proposed method achieves superior results compared to the state-of-the-art models in the full training and low-re-sourcesettings.

关键词： Medical dialogue understanding information extraction text generation knowledge-enhanced prompt low-resource setting dataaugmentation

Comprehensive Relation Modelling for Image Paragraph Generation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Machine Intelligence research 2024年第2期21卷 369-382页

作者： Xianglu Zhu Zhang Zhang Wei Wang Zilei Wang automation Department University of Science and Technology of ChinaHefei 230027China Center for research on Intelligent Perception and Computing National Laboratory of Pattern RecognitionInstitute of AutomationChinese Academy of SciencesBeijing 100190China University of Chinese academy of sciences Beijing 100864China

Image paragraph generation aims to generate a long description composed of multiple sentences,which is different from traditional image captioning containing only one *** of previous methods are dedicated to extracting rich features from image regions,and ignore modelling the visual *** this paper,we propose a novel method to generate a paragraph by modelling visual relationships ***,we parse an image into a scene graph,where each node represents a specific object and each edge denotes the relationship between two ***,we enrich the object features by implicitly encoding visual relationships through a graph convolutional network(GCN).We further explore high-order relations between different relation features using another graph convolutional *** addition,we obtain the linguistic features by projecting the predicted object labels and their relationships into a semantic embedding *** these features,we present an attention-based topic generation network to select relevant features and produce a set of topic vectors,which are then utilized to generate multiple *** evaluate the proposed method on the Stanford image-paragraph dataset which is currently the only available dataset for image paragraph generation,and our method achieves competitive performance in comparison with other state-of-the-art(SOTA)methods.

关键词： Image paragraph generation visual relationship scene graph graph convolutional network(GCN) long short-term memory

Underwater RGB-D imaging system with millimetric precision

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science china(Information sciences) 2024年第10期67卷 383-385页

作者： Yajun GAO Yang CONG Xu TANG Mingxue LI State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences University of Chinese academy of sciences College of automation Science and Engineering South China University of Technology

In recent years, underwater fine three-dimensional reconstruction technology has emerged as a focal point in research, exhibiting extensive potential applications, particularly in underwater archaeology and resource e... 详细信息

关键词： SINGLE CAMERA

同方期刊数据库

A flat-foldable equiangular spiral folding pattern inspired by sunflowers for deployable structures

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第6期37卷 425-438页

作者： Shuangqing YU Jinguo LIU Pengyuan ZHAO Yingying TANG State Key Laboratory of Robotics Shenyang Institute of AutomationChinese Academy of SciencesShenyang 110016China institutes for Robotics and Intelligent Manufacturing Chinese Academy of SciencesShenyang 110169China University of Chinese academy of sciences Beijing 100049China

Flasher origami pattern has been widely utilized to improve the stowage efficiency of deployable ***,flasher origami cannot be folded fully flat,and they still have great potential for optimization in terms of storage volume and folding *** this paper,a flat foldable equiangular spiral folding pattern inspired by the sunflower disk is ***,a parametric design method for this equiangular spiral crease diagram is introduced in ***,a kinematic model of the equiangular spiral folding pattern is established based on the kinematic equivalence between rigid origami and spherical linkages.A simulation of the developed model demonstrates that the equiangular spiral folding pattern can be folded *** the folded ratio as an evaluation index,the calculated results and experiments show that the equiangular spiral crease pattern can yield fewer creases and improve stowage efficiency in comparison to flasher origami *** spiral folding pattern can save a considerable amount of space and provide a new approach to spatially deployable structures.

关键词： Flasher origami Sunflower-inspired Deployable structures Flat-foldability Equiangular spiral folding pattern

Fixed-Time Antidisturbance Consensus Tracking for Nonlinear Multiagent Systems With Matching and Mismatching Disturbances

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第6期11卷 1410-1423页

作者： Xiangmin Tan Chunyan Hu Guanzhen Cao Qinglai Wei Wei Li Bo Han institute of Engineering Thermophysics Chinese Academy of SciencesBeijing 100190 School of Aeronautics and Astronautics University of Chinese Academy of SciencesBeijing 100049China institute of automation Chinese Academy of SciencesBeijing 100190 School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China IEEE

In this paper,fixed-time consensus tracking for mul-tiagent systems(MASs)with dynamics in the form of strict feed-back affine nonlinearity is addressed.A fixed-time antidistur-bance consensus tracking protocol is proposed,which consists of a distributed fixed-time observer,a fixed-time disturbance observer,a nonsmooth antidisturbance backstepping controller,and the fixed-time stability analysis is conducted by using the Lyapunov theory *** paper includes three main ***,a distributed fixed-time observer is developed for each follower to obtain an estimate of the leader’s output by utilizing the topology of the communication ***,a fixed-time disturbance observer is given to estimate the lumped disturbances for feedforward ***,a nonsmooth antidisturbance backstepping tracking controller with feedforward compensation for lumped disturbances is *** order to mitigate the“explosion of complexity”in the tradi-tional backstepping approach,we have implemented a modified nonsmooth command filter to enhance the performance of the closed-loop *** simulation results show that the pro-posed method is effective.

关键词： Antidisturbance backstepping consensus tracking fixed-time stability multiagent system(MASs) strict feedback affine nonlinear systems.

Vision-based docking system for an aromatic-hydrocarbon-inspired reconfigurable robot

在线全文

学校读者我要写书评

暂无评论

Science china(Technological sciences) 2024年第6期67卷 1798-1816页

作者： XING FangYi XU Cheng LIU JinGuo XUE ZhiHui State Key Laboratory of Robotics Shenyang Institute of AutomationChinese Academy of SciencesShenyang110016China School of automation and Electrical Engineering Shenyang Ligong UniversityShenyang110158China institutes for Robotics and Intelligent Manufacturing Chinese Academy of SciencesShenyang110169China University of Chinese academy of sciences Beijing100049China

Aromatic hydrocarbons generally refer to compounds containing benzene *** types of isomers can be formed by replacing hydrogen atoms on the benzene *** this paper,an aromatic-hydrocarbon-inspired modular robot(AHIMR)is *** robot can be reassembled into different configurations suitable for various task requirements.A vision-based docking system is designed for the *** system primarily consists of two stages:a remote guidance stage and a precise docking *** the remote guidance stage,an object module is identified using an illumination adaptive target recognition algorithm,and then the active module moves to the docking area through communication with *** the precise docking stage,the active module calculates the relative pose with the object module using a perspective-n-point method and dynamically adjusts its posture to *** this process,a Kalman filter is used to reduce target occlusion and jitter *** addition,the docking system feasibility is verified via several simulation *** module docking accuracy is controlled within 0.01 m,which meets the reconfiguration task requirements of the *** the AHIMR submodule docking experiment,the active module accurately moves to the expected position with a docking success rate of 95%.

关键词： modular robot reconfiguration aromatic hydrocarbon autonomous docking visual identification

Corporate Credit Ratings Based on Hierarchical Heterogeneous Graph Neural Networks

在线全文

学校读者我要写书评

暂无评论

Machine Intelligence research 2024年第2期21卷 257-271页

作者： Bo-Jing Feng Xi Cheng Hao-Nan Xu Wen-Fang Xue Center for research on Intelligent Perception and Computing National Laboratory of Pattern RecognitionInstitute of AutomationChinese Academy of SciencesBeijing 100190China

order to help investors understand the credit status of target corporations and reduce investment risks,the corporate credit rating model has become an important evaluation tool in the financial *** models are based on statistical learning,machine learning and deep learning especially graph neural networks(GNNs).However,we found that only few models take the hierarchy,heterogeneity or unlabeled data into account in the actual corporate credit rating ***,we propose a novel framework named hierarchical heterogeneous graph neural networks(HHGNN),which can fully model the hierarchy of corporate features and the heterogeneity of relationships between *** addition,we design an adversarial learning block to make full use of the rich unlabeled samples in the financial *** experiments conducted on the public-listed corporate rating dataset prove that HHGNN achieves SOTA compared to the baseline methods.

关键词： Corporate credit rating hierarchical relation heterogeneous graph neural networks adversarial learning

Multimodal Fusion of Brain Imaging Data: Methods and Applications

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Machine Intelligence research 2024年第1期21卷 136-152页

作者： Na Luo Weiyang Shi Zhengyi Yang Ming Song Tianzi Jiang Brainnetome Center and National Laboratory of Pattern Recognition Institute of AutomationChinese Academy of SciencesBeijing 100190China School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China Center for Excellence in Brain Science and Intelligence Technology Institute of AutomationChinese Academy of SciencesBeijing 100190China research Center for Augmented Intelligence Zhejiang LaboratoryHangzhou 311100China

Neuroimaging data typically include multiple modalities,such as structural or functional magnetic resonance imaging,dif-fusion tensor imaging,and positron emission tomography,which provide multiple views for observing and analyzing the *** lever-age the complementary representations of different modalities,multimodal fusion is consequently needed to dig out both inter-modality and intra-modality *** the exploited rich information,it is becoming popular to combine multiple modality data to ex-plore the structural and functional characteristics of the brain in both health and disease *** this paper,we first review a wide spectrum of advanced machine learning methodologies for fusing multimodal brain imaging data,broadly categorized into unsupervised and supervised learning *** by this,some representative applications are discussed,including how they help to under-stand the brain arealization,how they improve the prediction of behavioral phenotypes and brain aging,and how they accelerate the biomarker exploration of brain ***,we discuss some exciting emerging trends and important future ***,we intend to offer a comprehensive overview of brain imaging fusion methods and their successful applications,along with the chal-lenges imposed by multi-scale and big data,which arises an urgent demand on developing new models and platforms.

关键词： Multimodal fusion supervised learning unsupervised learning brain atlas cognition brain disorders