Pose variation is one of the challenging factors for face recognition. In this paper, we propose a novel cross-pose face recognition method named as Regularized Latent Least Square Regression (RLLSR). The basic assump...
详细信息
ISBN:
(纸本)9781577356332
Pose variation is one of the challenging factors for face recognition. In this paper, we propose a novel cross-pose face recognition method named as Regularized Latent Least Square Regression (RLLSR). The basic assumption is that the images captured under different poses of one person can be viewed as pose-specific transforms of a single ideal object. We treat the observed images as regressor, the ideal object as response, and then formulate this assumption in the least square regression framework, so as to learn the multiple pose-specific transforms. Specifically, we incorporate some prior knowledge as two regularization terms into the least square approach: 1) the smoothness regularization, as the transforms for nearby poses should not differ too much;2) the local consistency constraint, as the distribution of the latent ideal objects should preserve the geometric structure of the observed image space. We develop an alternating algorithm to simultaneously solve for the ideal objects of the training individuals and a set of pose-specific transforms. The experimental results on the Multi-PIE dataset demonstrate the effectiveness of the proposed method and superiority over the previous methods.
A docking approach for child robot suitable for marsupial robot was designed in this paper. The mother robot obtained the posture of child robot relative to the docking entrance by visual perception to instruct the ch...
详细信息
A docking approach for child robot suitable for marsupial robot was designed in this paper. The mother robot obtained the posture of child robot relative to the docking entrance by visual perception to instruct the child robot. In each of left, right and back sides of child robot, there was a color symbol with two colors arranged vertically. The mother robot extracted the color borderline of each symbol it observed in a way of connection growing, and then the position information of corresponding color symbol in image as well as the length ratio could be acquired to guide the motion of child robot. Finally, the child robot entered the docking station in an appropriate posture. This approach was verified by experimental results.
One of the most difficult challenges in automatic face recognition is computing facial similarity between two images captured in different modalities, called heterogeneous face recognition. In this paper, we propose a...
详细信息
A vast amount of complex spatio-temporal brain data, such as EEG-, have been accumulated. Technological advances in many disciplines rely on the proper analysis, understanding and utilisation of these data. In order t...
详细信息
This paper discusses the energy minimization problem of a class of chaotic systems, and constructs an optimal neuro-controller based on adaptive dynamic programming (ADP) algorithm. To learn the optimal performance in...
详细信息
An object grasping method based on fuzzy approaching for a mobile manipulator with an Eye-in-Hand CMOS (complementary metal-oxide-semiconductor transistor) camera was proposed. The approaching guidance identifier with...
详细信息
An object grasping method based on fuzzy approaching for a mobile manipulator with an Eye-in-Hand CMOS (complementary metal-oxide-semiconductor transistor) camera was proposed. The approaching guidance identifier with double cross was recognized by the mobile manipulator, and the angle between the line from mobile platform to the center of the identifier and its heading direction were served as the inputs of fuzzy controller. A double input and single output fuzzy controller was designed to regulate the direction of the mobile platform for smooth approaching. When the object (a red cylinder with double black lines) was in the workspace of manipulator, the mobile platform stops and the object was grasped by the manipulator with its joint angles solved based on inverse kinematics. Experiments results show the validity of the proposed approach.
A motion control method for mobile robot based on sub-regions evaluation in local sensing environment was proposed. Firstly, the local environment was divided into multiple sub-regions based on the sensing information...
详细信息
A motion control method for mobile robot based on sub-regions evaluation in local sensing environment was proposed. Firstly, the local environment was divided into multiple sub-regions based on the sensing information provided by laser range finder. Then, each sub-region was evaluated by considering the distance influence factor, as well as visual guide and memory-based judgment. Finally, the optimal sub-region adapted to current environment will be used for motion decision of mobile robot. Experimental results show that the proposed method is capable of dividing sensing environment into multiple sub-regions and producing the optimal sub-region adapted to current environment. Motion control for robot was realized and the effectiveness of the proposed method was verified.
A robotic fish propelling itself by two long fins was designed and implemented from the inspiration of stingrays. The yaw control of this robotic fish was studied. The dynamic model was studied based on the stress ana...
详细信息
A robotic fish propelling itself by two long fins was designed and implemented from the inspiration of stingrays. The yaw control of this robotic fish was studied. The dynamic model was studied based on the stress analysis of the robotic fish. A yaw controller was designed based on the active disturbance rejection control (ADRC) technique. The model error and external disturbance were unified into a total system disturbance, which was estimated by an extended state observer. By compensating the total system disturbance in the control signal, the yaw control system was simplified to a typical second order cascade integral system which could be easily controlled by use of a more effective nonlinear feedback control. The ADRC based yaw controller was verified through simulations showing that the designed controller can effectively control the yaw angle of the robotic fish and has good dynamic and static characteristics.
One of the most important issues among active rehabilitation technique is how to extract the voluntary intention of patient through bio-signals, especially EEG signal. This pilot study investigates the feasibility of ...
详细信息
In this paper, a novel adaptive dynamic programming algorithm based on policy iteration is developed to solve online multi-player non-zero-sum differential game for continuous-time nonlinear systems. This algorithm is...
详细信息
In this paper, a novel adaptive dynamic programming algorithm based on policy iteration is developed to solve online multi-player non-zero-sum differential game for continuous-time nonlinear systems. This algorithm is mathematically equivalent to the quasi-Newton's iteration in a Banach space. The implementation using neural networks is given, where a critic neural network is used to learn its value function, and an action neural network sharing the same parameters with the corresponding critic neural network is used to learn its optimal control policy for each player. All the critic and action neural networks are updated online in real-time and continuously. A simulation example is presented to demonstrate the effectiveness of the developed scheme.
暂无评论