In this paper, we present a simple yet effective visual tracking algorithm with an appearance model based on 2D discrete cosine transform (2D-DCT) representations. The DCT has the properties of decorrelation and energ...
详细信息
ISBN:
(纸本)9781479973989
In this paper, we present a simple yet effective visual tracking algorithm with an appearance model based on 2D discrete cosine transform (2D-DCT) representations. The DCT has the properties of decorrelation and energy compaction, and is robust against geometry and illumination changes. Hence, it is suitable for appearance modeling and the features of our appearance model are extracted from an optimized low dimensional subspace. In order to adapt to the appearance change caused by environment change or ego motion, we also propose to update the observation appearance model through a nonlinear weighted method. Numerous experiments on some challenging video sequences demonstrated that our algorithm is effective and it considerably outperforms the other methods.
With the continuous development of data storage technology, the complexity, category and size of data increase sharply; on the other hand, the wide expansion of its application domain brings great challenges on the tr...
详细信息
ISBN:
(纸本)9781479960590
With the continuous development of data storage technology, the complexity, category and size of data increase sharply; on the other hand, the wide expansion of its application domain brings great challenges on the traditional data processing and analysis technology. In this paper, the data analysis and processing technology and its development status are introduced and by multi-dimensional data visualization analysis method and combining the knowledge of domain experts, the equipment purchasing fund data of a large enterprise is analyzed to provide reliable management decision to the investment and investment volume of business fund.
With the ubiquity of mobile communication devices,people experiencing traffic jams share real-time information and interact with each other on social media sites,which provide new channels to monitor,estimate and mana...
With the ubiquity of mobile communication devices,people experiencing traffic jams share real-time information and interact with each other on social media sites,which provide new channels to monitor,estimate and manage traffic *** this paper,we use natural language processing and data mining technologies to extract traffic jam related information from ***,analyze the content of people’s talk to discover the "talking point" of people when facing traffic jams,and to provide data support for relevant authorities to make successful and effective decisions for real-time traffic jam response and management.
Seamless integration of 3D acquisition and 3D display systems offers enhanced experience in 3D visualization of the real world objects or scenes. The vivid representation of captured 3D objects displayed on a glasses-...
详细信息
ISBN:
(纸本)9780819498922
Seamless integration of 3D acquisition and 3D display systems offers enhanced experience in 3D visualization of the real world objects or scenes. The vivid representation of captured 3D objects displayed on a glasses-free 3D display screen could bring the realistic viewing experience to viewers as if they are viewing real-world scene. Although the technologies in 3D acquisition and 3D display have advanced rapidly in recent years, effort is lacking in studying the seamless integration of these two different aspects of 3D technologies. In this paper, we describe our recent progress on integrating a light-field 3D acquisition system and an autostereoscopic multiview 3D display for real-time light field capture and display. This paper focuses on both the architecture design and the implementation of the hardware and the software of this integrated 3D system. A prototype of the integrated 3D system is built to demonstrate the real-time 3D acquisition and 3D display capability of our proposed system.
A mother or transporting robot is designed in this paper, which is dedicated to retrieve, transport and deploy the children or smaller robots to configure a robotic team called the marsupial robotic system. In order t...
详细信息
A mother or transporting robot is designed in this paper, which is dedicated to retrieve, transport and deploy the children or smaller robots to configure a robotic team called the marsupial robotic system. In order to manage children robots flexibly, a multi-floor docking station is mounted on the mother robot with a lifting platform which is dragged via lead-screw driving. Touch switches are utilized to initialize the lifting height and identify whether a child robot arrives at the parking position. Also, a dock camera is set on the top of the station to recognize different children robots for the purpose of deploying or retrieving. The locomotion ability is enhanced by combining the advantages of the wheeled-mobile platform and tracked-mobile platform with the concept of wheel-track combo, which barely increases the complexity of mechanism. Finally, a prototype of mother robot is developed and implemented.
Recently, attributes have been introduced as a kind of high-level semantic information to help improve the classification accuracy. Multitask learning is an effective methodology to achieve this goal, which shares low...
详细信息
Recently, attributes have been introduced as a kind of high-level semantic information to help improve the classification accuracy. Multitask learning is an effective methodology to achieve this goal, which shares low-level features between attributes and actions. Yet such methods neglect the constraints that attributes impose on classes, which may fail to constrain the semantic relationship between the attributes and actions. In this paper, we explicitly consider such attribute-action relationship for human action recognition, and correspondingly, we modify the multitask learning model by adding attribute regularization. In this way, the learned model not only shares the low-level features, but also gets regularized according to the semantic constrains. In addition, since attribute and class label contain different amounts of semantic information, we separately treat attribute classifiers and action classifiers in the framework of multitask learning for further performance improvement. Our method is verified on three challenging datasets (KTH, UIUC, and Olympic Sports), and the experimental results demonstrate that our method achieves better results than that of previous methods on human action recognition.
Understanding the strategies to optimize/suppress information spreads under intense competition could provide important insights in a broad range of settings including viral marketing,emergency response and informatio...
详细信息
Understanding the strategies to optimize/suppress information spreads under intense competition could provide important insights in a broad range of settings including viral marketing,emergency response and information system ***,most of existing studies about competitive influence diffusion mainly focus on two-information competition *** date,the competitive influence maximization problem considering the mechanism of multi-information competition is still not well *** this paper,we conducted computational experiments to study the competitive influence maximization with multi-information competition *** applying an information diffusion model called limited attention model(LAM),we carried on two computational experiments to validate the model and investigate the relation between seed selection methods and the properties of information *** experimental results show that 1)the LAM model could reproduce the features of empirical distribution in Chinese social media;2)the eigenvector centrality-based heuristic is a reasonable seed selection method for competitive influence maximization *** results of this paper can provide significant potential implications for information system design and management.
In this paper, aiming at the indoor scene under monitoring by visual sensor network (VSN), an object recognition approach based on structural feature is presented. Firstly, we regard the output of existing line segmen...
详细信息
In this paper, aiming at the indoor scene under monitoring by visual sensor network (VSN), an object recognition approach based on structural feature is presented. Firstly, we regard the output of existing line segment detector LSD with proper parameters as the preliminary extraction result and it still will be further restored and split. Then, we give an inference model based on structural features of object including line segment ontology characteristics and relative relationship between the line segments. Finally, the objects are recognized with position information through inference. The effectiveness of the approach is verified, and the results show that our approach does not rely on segmentation and has robustness on partial defect and structural deformation to some extent.
In this paper, a model-free and effective approach is proposed to solve infinite horizon optimal control problem for affine nonlinear systems based on adaptive dynamic programming technique. The developed approach, re...
详细信息
In this paper, a model-free and effective approach is proposed to solve infinite horizon optimal control problem for affine nonlinear systems based on adaptive dynamic programming technique. The developed approach, referred to as the actor-critic structure, employs two multilayer perceptron neural networks to approximate the state-action value function and the control policy, respectively. It uses data collected arbitrarily from any reasonable sampling distribution for policy iteration. In the policy evaluation phase, a novel objective function is defined for updating the critic network, and thus makes the critic network converge to the Bellman equation directly rather than iteratively. In the policy improvement phase, the action network is updated to minimize the outputs of the critic network. The two phases alternate until no more improvement of the control policy is observed, such that the optimal control policy is achieved. Two simulation examples are provided to show the effectiveness of the approach.
暂无评论