Image-text retrieval is a challenge topic since image features are still not good enough to represent the high-level semantic information, though the representation ability is improved thanks to advances in deep learning. This paper proposes a cross-modal image-text retrieval framework (BiTGRMAC) based on big transfer and region maximum activation convolution with generalized attention, where big transfer (BiT) trained with large amount data is utilized to extract image features and fine-tuned on the cross-modal image datasets. At the same time, a new generalized attention region maximum activation convolution (GRMAC) descriptor is introduced into BiT, which can generate image features through attention mechanism, then reduce the influence of background clustering and highlight the target. For texts, the widely used Sentence CNN is adopted to extract text features. The parameters of image and text deep models are learned by minimizing a cross-modal loss function in an end-to-end framework. Experimental results show that this method can effectively improve the accuracy of retrieval on three widely used datasets.
{"title":"Cross-modal Retrieval based on Big Transfer and Regional Maximum Activation of Convolutions with Generalized Attention","authors":"Wenwen Yang, Yan Hua","doi":"10.1145/3483845.3483872","DOIUrl":"https://doi.org/10.1145/3483845.3483872","url":null,"abstract":"Image-text retrieval is a challenge topic since image features are still not good enough to represent the high-level semantic information, though the representation ability is improved thanks to advances in deep learning. This paper proposes a cross-modal image-text retrieval framework (BiTGRMAC) based on big transfer and region maximum activation convolution with generalized attention, where big transfer (BiT) trained with large amount data is utilized to extract image features and fine-tuned on the cross-modal image datasets. At the same time, a new generalized attention region maximum activation convolution (GRMAC) descriptor is introduced into BiT, which can generate image features through attention mechanism, then reduce the influence of background clustering and highlight the target. For texts, the widely used Sentence CNN is adopted to extract text features. The parameters of image and text deep models are learned by minimizing a cross-modal loss function in an end-to-end framework. Experimental results show that this method can effectively improve the accuracy of retrieval on three widely used datasets.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123419953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the process of the rapid development of the information society, there are still many problems in the integrity, accuracy and security of image data. Blockchain technology stands out with its characteristics of decentralization, data immutability and data openness and transparency. Therefore, in view of the above problems, this paper proposes a block chain-based image copyright confirmation and protection model. The model includes five processes: consensus node joining, generation of registration data, consensus registration data, storage of registration data, and digital watermarking. Ensure that the image information cannot be tampered with after it is stored on the blockchain, and all nodes jointly maintain the storage; At the same time, the digital watermark is added to prevent the image information from leaking out of the chain to be embezzled by criminals for personal gain.
{"title":"Research on Image Copyright Confirmation and Protection Model Based on Blockchain","authors":"Zhengliang Wang, Taijun Li","doi":"10.1145/3483845.3483886","DOIUrl":"https://doi.org/10.1145/3483845.3483886","url":null,"abstract":"In the process of the rapid development of the information society, there are still many problems in the integrity, accuracy and security of image data. Blockchain technology stands out with its characteristics of decentralization, data immutability and data openness and transparency. Therefore, in view of the above problems, this paper proposes a block chain-based image copyright confirmation and protection model. The model includes five processes: consensus node joining, generation of registration data, consensus registration data, storage of registration data, and digital watermarking. Ensure that the image information cannot be tampered with after it is stored on the blockchain, and all nodes jointly maintain the storage; At the same time, the digital watermark is added to prevent the image information from leaking out of the chain to be embezzled by criminals for personal gain.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127568799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the era of new media, the spread of negative corporate public opinion in social networks has a significant influence on enterprises and society. In order to improve the ability of enterprises to respond to public opinion, it is important to study the evolutionary dissemination process of negative corporate public opinion. Firstly, BA network and ER network are used to simulate the online network and offline network. Then, we construct a model for the spread of negative corporate public opinion on both online and offline networks based on the real situation, and providing the corresponding mean field equation and solving the threshold value for the spread of negative corporate public opinion. Finally, the model is validated with the Samsung cell phone explosion incident and some response strategies are proposed based on image restoration theory. The results of the study show that the study of negative corporate opinion dissemination process in online and offline two-layer networks can better reflect the evolution of negative corporate opinion in real society than in single-layer networks; After the occurrence of negative public opinion events, enterprises should take appropriate response strategies in time, otherwise it will be counterproductive.
{"title":"The Communication Model of Negative Public Opinions of Corporate Based on Two-Layer Network","authors":"Shu-qiong Chen, Xiaoli Wang","doi":"10.1145/3483845.3483868","DOIUrl":"https://doi.org/10.1145/3483845.3483868","url":null,"abstract":"In the era of new media, the spread of negative corporate public opinion in social networks has a significant influence on enterprises and society. In order to improve the ability of enterprises to respond to public opinion, it is important to study the evolutionary dissemination process of negative corporate public opinion. Firstly, BA network and ER network are used to simulate the online network and offline network. Then, we construct a model for the spread of negative corporate public opinion on both online and offline networks based on the real situation, and providing the corresponding mean field equation and solving the threshold value for the spread of negative corporate public opinion. Finally, the model is validated with the Samsung cell phone explosion incident and some response strategies are proposed based on image restoration theory. The results of the study show that the study of negative corporate opinion dissemination process in online and offline two-layer networks can better reflect the evolution of negative corporate opinion in real society than in single-layer networks; After the occurrence of negative public opinion events, enterprises should take appropriate response strategies in time, otherwise it will be counterproductive.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127198221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we propose an efficient algorithm to solve the weapon target assignment (WTA) problem combining the advantages of rule-based with that of traditional optimization methods. The main ideal of the proposed algorithm is building an adaptive neuro-fuzzy inference system (ANFIS) to obtain an original assignment scheme, and then the original scheme is used to initialize particles in discrete particle swarm optimization (DPSO). With the original assignment scheme provided by ANFIS, it can solve the problem of converging to local optimum with random initialization in DPSO efficiently. At last, a numerical simulation is proposed to illustrate the efficiency of the method in this paper.
{"title":"ANFIS-based Self-learning Expert System for Weapon Target Assignment Problem","authors":"Changcheng Wang, Lisi Chen, Wencai Li, Kan Zeng","doi":"10.1145/3483845.3483863","DOIUrl":"https://doi.org/10.1145/3483845.3483863","url":null,"abstract":"In this paper, we propose an efficient algorithm to solve the weapon target assignment (WTA) problem combining the advantages of rule-based with that of traditional optimization methods. The main ideal of the proposed algorithm is building an adaptive neuro-fuzzy inference system (ANFIS) to obtain an original assignment scheme, and then the original scheme is used to initialize particles in discrete particle swarm optimization (DPSO). With the original assignment scheme provided by ANFIS, it can solve the problem of converging to local optimum with random initialization in DPSO efficiently. At last, a numerical simulation is proposed to illustrate the efficiency of the method in this paper.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121228002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lilei Zhao, Liying Su, Nianqiang Cui, Zhuo Zhang, Xuping Zhang
Cell microinjection has been widely used in medical and biological research because of its high transmission efficiency and low toxicity. The overlarge deformation of cells during microinjection is one important potential reason for its low survival rate after injection. Therefore, there are the existing studies that have tried to established linear mechanic deformation models of biological cells under microinjection. In this paper, based on electrothermally driven microclamps, a nonlinear viscoelastic model is established to estimate the deformation of a Zebrafish embryo clamped by a microgripper during microinjection. Firstly, a nonlinear viscoelastic model is proposed to describe the dynamic behavior with respect to injection speed, cell deformation, and puncture force of Zebrafish embryos during injection. Secondly, microinjection experiments of Zebrafish embryos are conducted to measure the injection force and deformation at different injection speeds. Finally, the parameters of the nonlinear viscoelastic model are identified with the experimental results, and the proposed model in this paper is compared with the existing linear model in terms of accuracy.
{"title":"A Nonlinear Mechanic Model of a Zebrafish Embryo under Microinjection","authors":"Lilei Zhao, Liying Su, Nianqiang Cui, Zhuo Zhang, Xuping Zhang","doi":"10.1145/3483845.3483895","DOIUrl":"https://doi.org/10.1145/3483845.3483895","url":null,"abstract":"Cell microinjection has been widely used in medical and biological research because of its high transmission efficiency and low toxicity. The overlarge deformation of cells during microinjection is one important potential reason for its low survival rate after injection. Therefore, there are the existing studies that have tried to established linear mechanic deformation models of biological cells under microinjection. In this paper, based on electrothermally driven microclamps, a nonlinear viscoelastic model is established to estimate the deformation of a Zebrafish embryo clamped by a microgripper during microinjection. Firstly, a nonlinear viscoelastic model is proposed to describe the dynamic behavior with respect to injection speed, cell deformation, and puncture force of Zebrafish embryos during injection. Secondly, microinjection experiments of Zebrafish embryos are conducted to measure the injection force and deformation at different injection speeds. Finally, the parameters of the nonlinear viscoelastic model are identified with the experimental results, and the proposed model in this paper is compared with the existing linear model in terms of accuracy.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133840715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The goal of the system proposed in this paper is to develop functions for humanoid robots to use vision to perceive human behavior and imitate human motions. With that aim, the position of the human joint points is recorded by a key point detector and the motion data is fed to 3D-baseline to estimate the 3D human skeleton. The accuracy of 3D joint point prediction is directly related to the degree of robot's restoration of human motions. Therefore, data augmentation is applied to improve the accuracy of 3D human pose estimation. The experimental results on Human3.6M demonstrate that our method can yield good performance. In addition, the motion mapping between human and humanoid robot is realized by inverse kinematics. A balance strategy based on ankle joint adjustment is proposed to maintain the stability of the robot. Tested in a humanoid robot, the system is able to imitate human movements naturally after observation, which improves the human-computer interaction of the humanoid robot.
{"title":"Visual Perception Method Based on Human Pose Estimation for Humanoid Robot Imitating Human Motions","authors":"T. Tao, Zhiwei Zhang, Xingyu Yang","doi":"10.1145/3483845.3483894","DOIUrl":"https://doi.org/10.1145/3483845.3483894","url":null,"abstract":"The goal of the system proposed in this paper is to develop functions for humanoid robots to use vision to perceive human behavior and imitate human motions. With that aim, the position of the human joint points is recorded by a key point detector and the motion data is fed to 3D-baseline to estimate the 3D human skeleton. The accuracy of 3D joint point prediction is directly related to the degree of robot's restoration of human motions. Therefore, data augmentation is applied to improve the accuracy of 3D human pose estimation. The experimental results on Human3.6M demonstrate that our method can yield good performance. In addition, the motion mapping between human and humanoid robot is realized by inverse kinematics. A balance strategy based on ankle joint adjustment is proposed to maintain the stability of the robot. Tested in a humanoid robot, the system is able to imitate human movements naturally after observation, which improves the human-computer interaction of the humanoid robot.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117199945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Adaptive neural impedance control based on reference impedance model is introduced. Both model parameter uncertainties and model uncertainties are considered in controller design. The designed controller based reference impedance model ensure similar dynamics between robot and reference model. In order to handle model parameter uncertainties, the adaptive controller is designed and model uncertainties is estimated with neural network based radial basis function. System closed-loop stability is proved by Lyapunov theorem and the performance of proposed control method is verified by simulation with two-DOFs robot.
{"title":"Adaptive Neural Network Impedance Control of Robots Based on Reference Model","authors":"Ping Zhou, Wei-qing Ai, Longhe Yang","doi":"10.1145/3483845.3483858","DOIUrl":"https://doi.org/10.1145/3483845.3483858","url":null,"abstract":"Adaptive neural impedance control based on reference impedance model is introduced. Both model parameter uncertainties and model uncertainties are considered in controller design. The designed controller based reference impedance model ensure similar dynamics between robot and reference model. In order to handle model parameter uncertainties, the adaptive controller is designed and model uncertainties is estimated with neural network based radial basis function. System closed-loop stability is proved by Lyapunov theorem and the performance of proposed control method is verified by simulation with two-DOFs robot.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133241729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tanmay Vilas Samak, Chinmay Vilas Samak, Mingjuan Xie
AutoDRIVE is envisioned to be an integrated research and education platform for scaled autonomous vehicles and related applications. This work is a stepping-stone towards achieving the greater goal of realizing such a platform. Particularly, this work introduces the AutoDRIVE Simulator, a high-fidelity simulator for scaled autonomous vehicles. The proposed simulation ecosystem is developed atop the Unity game engine, and exploits its features in order to simulate realistic system dynamics and render photorealistic graphics. It comprises of a scaled vehicle model equipped with a comprehensive sensor suite for redundant perception, a set of actuators for constrained motion control and a fully functional lighting system for illumination and signaling. It also provides a modular environment development kit, which comprises of various environment modules that aid in reconfigurable construction of the scene. Additionally, the simulator features a communication bridge in order to extend an interface to the autonomous driving software stack developed independently by the users. This work describes some of the prominent components of this simulation system along with some key features that it has to offer in order to accelerate education and research aimed at autonomous driving.
{"title":"AutoDRIVE Simulator: A Simulator for Scaled Autonomous Vehicle Research and Education","authors":"Tanmay Vilas Samak, Chinmay Vilas Samak, Mingjuan Xie","doi":"10.1145/3483845.3483846","DOIUrl":"https://doi.org/10.1145/3483845.3483846","url":null,"abstract":"AutoDRIVE is envisioned to be an integrated research and education platform for scaled autonomous vehicles and related applications. This work is a stepping-stone towards achieving the greater goal of realizing such a platform. Particularly, this work introduces the AutoDRIVE Simulator, a high-fidelity simulator for scaled autonomous vehicles. The proposed simulation ecosystem is developed atop the Unity game engine, and exploits its features in order to simulate realistic system dynamics and render photorealistic graphics. It comprises of a scaled vehicle model equipped with a comprehensive sensor suite for redundant perception, a set of actuators for constrained motion control and a fully functional lighting system for illumination and signaling. It also provides a modular environment development kit, which comprises of various environment modules that aid in reconfigurable construction of the scene. Additionally, the simulator features a communication bridge in order to extend an interface to the autonomous driving software stack developed independently by the users. This work describes some of the prominent components of this simulation system along with some key features that it has to offer in order to accelerate education and research aimed at autonomous driving.","PeriodicalId":134636,"journal":{"name":"Proceedings of the 2021 2nd International Conference on Control, Robotics and Intelligent System","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114106997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：