In one-stage methods for video moment retrieval,the common representations indirectly supervised by boundary prediction fail to fully preserve the inherent characteristic of the video and query,which limits the retrie...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
In one-stage methods for video moment retrieval,the common representations indirectly supervised by boundary prediction fail to fully preserve the inherent characteristic of the video and query,which limits the retrieval *** solve this problem,an Adversarial Video Moment Retrieval(AVMR) algorithm is proposed to learn the common representations with modality invariance and cross-modal *** is implemented through the process of adversarial learning between a feature projector and a modality *** feature projector tries to generate a modality-invariant common representation and to confuse the modality *** modality classifier tries to discriminate between different modalities based on the generated representation by the feature *** triplet constraints are further imposed on the feature projector to preserve the underlying cross-modal semantic structure of *** experimental results show that AVMR surpasses the baseline Attentive Cross-modal Relevance Matching(ACRM) by 1.10% and 1.73% in the "mIoU" metric on two public datasets Charades-STA and TACoS,respectively.
It is a tendency for Internet of Things to research Zigbee-WiFi hybrid *** paper introduces a heterogeneous network integration module based on protocol conversion which can integrate 802.15.4 network with 802.11 b **...
详细信息
ISBN:
(纸本)9781509009107
It is a tendency for Internet of Things to research Zigbee-WiFi hybrid *** paper introduces a heterogeneous network integration module based on protocol conversion which can integrate 802.15.4 network with 802.11 b *** heterogeneous network integration module resolves mismatch problems on physical layer,MAC layer and address mapping,meanwhile it makes a contrast with existing Zigbee-WiFi gateway and found to be advanced and effective in three performance indexes:network connection speed,SINR(signal interference noise ratio) and network *** design and realization of heterogeneous network integration are based on QualNet simulation software.
Unmanned systems are increasingly adopted in various fields,becoming an indispensable technology in daily *** systems are the lifeblood of unmanned systems,and affect the working time and task ***,traditional power sy...
详细信息
Unmanned systems are increasingly adopted in various fields,becoming an indispensable technology in daily *** systems are the lifeblood of unmanned systems,and affect the working time and task ***,traditional power systems,such as batteries and fuels have a fixed ***,once the power supply is exhausted and cannot be replenished in time,the unmanned systems will stop ***,researchers have increasingly begun paying attention to renewable energy generation *** principles,advantages,and limitations of renewable energy generation technologies are different,and their application effects in different unmanned systems are also *** paper presents a comprehensive study of the application and development status of photovoltaic,thermoelectric,and magnetoelectric generation technologies in four kinds of unmanned systems,including space,aviation,ground,and water,and then summarizes the adaptability and limitations of the three technologies to different ***,future development directions are predicted to enhance the reliability of renewable energy generation technologies in unmanned *** is the first study to conduct a comprehensive and detailed study of renewable energy generation technologies applied in unmanned *** present work is critical for the development of renewable energy generation technologies and power systems for unmanned systems.
For conventional adaptive control, time-varying parametric uncertainty and unmodeled dynamics are ticklish problems, which will lead to undesirable performance or even instability and nonrobust behavior, respectively....
详细信息
For conventional adaptive control, time-varying parametric uncertainty and unmodeled dynamics are ticklish problems, which will lead to undesirable performance or even instability and nonrobust behavior, respectively. In this study, a class of discrete-time switched systems with unmodeled dynamics is taken into consideration. Moreover, nonlinear systems are here supposed to be approximated with the class of switched systems considered in this paper, and thereby switching control design is investigated for both switched systems and nonlinear systems to assure stability and performance. For robustness against unmodeled dynamics and uncertainty, robust model reference adaptive control(RMRAC) law is developed as the basis of controller design for each individual subsystem in the switched systems or nonlinear systems. Meanwhile, two different switching laws are presented for switched systems and nonlinear systems, respectively. Thereby, the authors incorporate the corresponding switching law into the RMRAC law to construct two schemes of switching control respectively for the two kinds of controlled systems. Both closed-loop analyses and simulation examples are provided to illustrate the validity of the two proposed switching control schemes. Furthermore, as to the proposed scheme for nonlinear systems, its potential for practical application is demonstrated through simulations of longitudinal control for F-16 aircraft.
A facial expression emotion recognition based human-robot interaction(FEER-HRI) system is proposed, for which a four-layer system framework is designed. The FEERHRI system enables the robots not only to recognize huma...
详细信息
A facial expression emotion recognition based human-robot interaction(FEER-HRI) system is proposed, for which a four-layer system framework is designed. The FEERHRI system enables the robots not only to recognize human emotions, but also to generate facial expression for adapting to human emotions. A facial emotion recognition method based on2D-Gabor, uniform local binary pattern(LBP) operator, and multiclass extreme learning machine(ELM) classifier is presented,which is applied to real-time facial expression recognition for robots. Facial expressions of robots are represented by simple cartoon symbols and displayed by a LED screen equipped in the robots, which can be easily understood by human. Four scenarios,i.e., guiding, entertainment, home service and scene simulation are performed in the human-robot interaction experiment, in which smooth communication is realized by facial expression recognition of humans and facial expression generation of robots within 2 seconds. As a few prospective applications, the FEERHRI system can be applied in home service, smart home, safe driving, and so on.
Compressive sensing is a revolutionary idea proposed recently to achieve much lower sampling rate for *** the image application with limited resources the camera data can be stored and processed in compressed *** algo...
详细信息
Compressive sensing is a revolutionary idea proposed recently to achieve much lower sampling rate for *** the image application with limited resources the camera data can be stored and processed in compressed *** algorithm for moving object and region detection in video using a compressive sampling is *** algorithm estimates motion information of the moving object and regions in the video from the compressive measurements of the current image and background *** algorithm does not perform inverse compressive operation to obtain the actual pixels of the current image nor the estimated *** leads to a computationally efficient method and a system compared with the existing motion estimation *** experimental results show that the sampling rate can reduce to 25% without sacrificing performance.
The positioning accuracy of a short-haul target-locating system,the inverse-GPS(IGPS) ,was analyzed in detail. The relationship between IGPS and the positioning error was discussed. The multiplicative error minimal bo...
详细信息
The positioning accuracy of a short-haul target-locating system,the inverse-GPS(IGPS) ,was analyzed in detail. The relationship between IGPS and the positioning error was discussed. The multiplicative error minimal bound of the geometric dilution of precision (GDOP) about the four-base-station IGPS was also investigated. In order to clarify the practical implementation of IGPS,the multiplicative and additive error factors which affect the positioning accuracy and theoretical estimation of positioning accuracy were presented. By analyzing the experiments of locating a target's position in virtual three-dimensional areas,the positioning performance of IGPS was illustrated. The results show that the multiplicative and additive error factors should be eliminated in IGPS to improve the positioning accuracy.
This paper considers distributed optimization for minimizing the average of local nonconvex cost functions, by using local information exchange over undirected communication networks. To reduce the required communicat...
详细信息
Distributed stereoscopic rotating formation control of networks of second-order agents is investigated. A distributed control protocol is proposed to enable all agents to form a stereoscopic formation and surround a c...
详细信息
Distributed stereoscopic rotating formation control of networks of second-order agents is investigated. A distributed control protocol is proposed to enable all agents to form a stereoscopic formation and surround a common axis. Due to the existence of the rotating mode, the desired relative position between every two agents is time-varying, and a Lyapunov-based approach is employed to solve the rotating formation control problem. Finally, simulation results are provided to illustrate the effectiveness of the theoretical results.
In this work, we propose a new scheme to estimate the algebraic connectivity of the graph describing the network topology of a multi-agent system. We consider network topologies modeled by undirected graphs. The main ...
详细信息
暂无评论