In recent years,the introduction of Siamese network has brought new vitality to the object tracking ***,high-performance Siamese trackers cannot run at a real-time speed on mobile devices due to their complex and huge...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
In recent years,the introduction of Siamese network has brought new vitality to the object tracking ***,high-performance Siamese trackers cannot run at a real-time speed on mobile devices due to their complex and huge *** distillation is a common and effective model compression method,but it is difficult to be applied to the challenging task like object *** find out the fundamental cause is that the imbalance between the foreground and background in the object tracking task,which aggravates the problem of insufficient feature extraction ability of small ***,we propose the attention mask distillation(AMD) to help the student tracker focus on the foreground area faster and more *** attention mask can be easily obtained from the feature maps and brings fine-granularity to the traditional binary *** experimental results on OTB100 and VOT2018 show that our method enables the student tracker perform as well as the teacher *** the same time,it's able to run on the CPU at a hyper-real-time of 66 fps and achieves nearly 9 times model compression *** low computational and storage costs make it possible to deploy high-performance trackers on resource-constrained platforms.
Welcome to the new issue of the IEEE Transactions on Computational Social systems (TCSS). As the new Editor-in-Chief (EiC), I would like to take this opportunity to brief all of you about the state of our journal and ...
Welcome to the new issue of the IEEE Transactions on Computational Social systems (TCSS). As the new Editor-in-Chief (EiC), I would like to take this opportunity to brief all of you about the state of our journal and my plan for its future development and direction in the next few years.
In this paper,we propose a clique-based sparse reinforcement learning(RL) algorithm for solving cooperative *** aim is to accelerate the learning speed of the original sparse RL algorithm and to make it applicable for...
详细信息
In this paper,we propose a clique-based sparse reinforcement learning(RL) algorithm for solving cooperative *** aim is to accelerate the learning speed of the original sparse RL algorithm and to make it applicable for tasks decomposed in a more general ***,a transition function is estimated and used to update the Q-value function,which greatly reduces the learning ***,it is more reasonable to divide agents into cliques,each of which is only responsible for a specific *** this way,the global Q-value function is decomposed into the sum of several simpler local Q-value *** decomposition is expressed by a factor graph and exploited by the general maxplus algorithm to obtain the greedy joint *** results show that the proposed approach outperforms others with better performance.
Recently, there has been an increased interest in the use of social media data as important traffic information sources. In this paper, we review social media based transportation research with social network analysis...
详细信息
The rapid and orderly evacuation of passengers at the railway hub station in case of emergencies is an important issue for railway safety and efficiency. In this paper, a robot-guided passenger evacuation method is pr...
详细信息
The rapid and orderly evacuation of passengers at the railway hub station in case of emergencies is an important issue for railway safety and efficiency. In this paper, a robot-guided passenger evacuation method is proposed to help passengers search evacuation paths and avoid potential risks. The number and initial positions of robots are determined by using a k-means clustering approach. The exit assignment and evacuation paths of robots are calculated by using a hybrid bi-level optimization approach taking into account the cooperative mechanism between robots. Then, a robot-guided crowd evacuation dynamical model is built based on a modified social force model, in which a navigation force is introduced to influence the speed and direction of evacuees. A case study of a typical railway hub station is used to demonstrate the effectiveness of the proposed approach. The scenarios of the mall and platform are designed to verify the evacuation efficiency under different robot distribution schemes. The experimental results prove that setting up robots can effectively reduce evacuation time, and the utilization of exits is more balanced. The proposed optimal scheme shows the best performance in evacuation efficiency, including evacuation time and exit utilization rate, compared to the uniform distribution scheme and no robot scheme.
Fisher vector (FV), which could be seen as a bag of visual words (BOW) that encodes not only word counts but also higher-order statistics, works well with linear classifiers and has shown promising performance for ima...
详细信息
Fisher vector (FV), which could be seen as a bag of visual words (BOW) that encodes not only word counts but also higher-order statistics, works well with linear classifiers and has shown promising performance for image categorization. For character recognition, although standard BOW has been applied, the results are still not satisfactory. In this paper, we apply Fisher vector derived from Gaussian Mixture Models (GMM) based visual vocabularies on character recognition and integrate spatial information as well. We, give a comprehensive evaluation of Fisher vector with linear classifier on a series of challenging English and digits character recognition datasets, including both the handwritten and scene character recognition ones. Moreover, we also collect two Chinese scene character recognition datasets to evaluate the suitability of Fisher vector to represent Chinese characters. Through extensive experiments we make three contributions: (1) we demonstrate that FV with linear classifier could outperform most of the state-of-the-art methods for character recognition, even the CNN based ones and the superiority is more obvious when training samples are insufficient to train the networks;(2) we show that additional spatial information is very useful for character representation, especially for Chinese ones, which have more complex structures;and (3) the results also imply the potential of FV to represent new unseen categories, which is quite inspiring since it is quite difficult to collect enough training samples for large-category Chinese scene characters. (C) 2017 Elsevier Ltd. All rights reserved.
Similar cases recommendation is more and more popular in the internet inquiry. There have been lots of cases which have been solved perfectly, and recommending them to similar inquiries can not only save the patients&...
详细信息
ISBN:
(纸本)9781467371896
Similar cases recommendation is more and more popular in the internet inquiry. There have been lots of cases which have been solved perfectly, and recommending them to similar inquiries can not only save the patients' waiting time, but also giving more good references. However, the inquiry platform cannot understand the diversity of description, i.e. the same meaning with different description. This may shield some cases with very high quality answers. In this paper, based on deep learning, we proposed a retrieval model combining word embedding with language models. We use word embedding to solve the problem of description diversity, and then recommend the similar cases for the inquiries. The experiments are based on the data from ***, and the results show that our methods outperform the state-of-art methods.
In this paper, we present a novel and robust visual tracking algorithm to obtain accurate target position for a mobile robot following control. To identify and localize tracked target in consecutive frames, the spatio...
详细信息
In this paper, we present a novel and robust visual tracking algorithm to obtain accurate target position for a mobile robot following control. To identify and localize tracked target in consecutive frames, the spatio-temporal relationship between the object of interest and its local context is formulated by using the Bayesian framework, and the best target location is ascertained by computing a confidence map which maximizes an object location likelihood. Specifically, convolution operation on middle-level feature space is utilized to measure the similarity between the target and its surrounding regions, and convolution theorem is applied to speed up detecting and locating the tracked target. Based on the proposed tracking algorithm, a robust target tracker is designed for a mobile robot to estimate the image position of the target. In conjunction with the depth information captured by Microsoft Kinect and typical Proportional, Integral, Derivative control method for the mobile robot, the robust target tracking and following system is developed to integrate tracking accuracy and agile following. Extensive experiments on the tracking benchmark illustrate the impressive performance of our tracker. Despite illumination changes and partial occlusion, several real-life tests are executed on mobile robot platform pretty well.
作者：
Wang, Fei-YueChinese Acad Sci
Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China
From January 1 to May 30, 2023, we have received 1214 submissions, averaging 8.09 submissions per day. Based on this SPD, we are on track to surpass our estimated 2500 submissions by November [1], [2]. This implies mo...
From January 1 to May 30, 2023, we have received 1214 submissions, averaging 8.09 submissions per day. Based on this SPD, we are on track to surpass our estimated 2500 submissions by November [1], [2]. This implies more work for processing and evaluation; therefore we need to speed up our effort to recruit more able and responsible Associate Editors and Reviewers. I need help from you in this task with great appreciation.
暂无评论