Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982710
W. Zeng, Long Zhou, Renhong Xu, Biao Li
For the issue of image denoising, in order to avoid the traditional multi-scale sparse representation methods, which used blocks of different sizes as a base function to represent image, the non-separable wavelets were taken. Their advantages included revealing the multi-scale structure, depicting the texture structure under different scales, and separating different directions and different types of singularity structure in a certain extent. Based on non-separable wavelets, a multi-scale sparse denoising model in the wavelet domain was we established, and then a collaboration sparse model for the sub-bands contained similar structures was designed to enhance the stability and accuracy of the sparse representation. The results show that the denoising effect based on new approach is obvious superior to the K-SVD algorithm.
{"title":"Multi-scale sparse denoising model based on non-separable wavelet","authors":"W. Zeng, Long Zhou, Renhong Xu, Biao Li","doi":"10.1109/SPAC.2014.6982710","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982710","url":null,"abstract":"For the issue of image denoising, in order to avoid the traditional multi-scale sparse representation methods, which used blocks of different sizes as a base function to represent image, the non-separable wavelets were taken. Their advantages included revealing the multi-scale structure, depicting the texture structure under different scales, and separating different directions and different types of singularity structure in a certain extent. Based on non-separable wavelets, a multi-scale sparse denoising model in the wavelet domain was we established, and then a collaboration sparse model for the sub-bands contained similar structures was designed to enhance the stability and accuracy of the sparse representation. The results show that the denoising effect based on new approach is obvious superior to the K-SVD algorithm.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"27 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114127408","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982668
Wen Liu, Zhijun Song
In this paper, an Internet Public Opinion Monitoring System based on Administration for Industry and Commerce is proposed. The paper first analyzes the design of the system. Then describes the system in detail as five modules, they are Internet Collection Module, Data Preprocessing Module, Public Opinion Analysis Module, Service Module and System Management Module.
{"title":"Design and implementation of an Internet Public Opinion Monitoring System","authors":"Wen Liu, Zhijun Song","doi":"10.1109/SPAC.2014.6982668","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982668","url":null,"abstract":"In this paper, an Internet Public Opinion Monitoring System based on Administration for Industry and Commerce is proposed. The paper first analyzes the design of the system. Then describes the system in detail as five modules, they are Internet Collection Module, Data Preprocessing Module, Public Opinion Analysis Module, Service Module and System Management Module.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"242 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120900211","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982695
Tao Zhou
We propose a novel approach for detecting keypoints invariant to scale changes based on M-wavelet theory. The theory description and detecting process of our approach are presented The comparative evaluation of different detectors shows our approach can provides a competent performance in rotation invariant, scale invariant, illumination invariant and noiseproof. In terms of scale changes, our proposed approach improves keypoint repeatability by 2%~10% compared with scale invariant feature transform (SIFT), speeded up robust features (SURF), Harris-Laplace, Hessian-Laplace.
{"title":"A scale invariant keypoints detector","authors":"Tao Zhou","doi":"10.1109/SPAC.2014.6982695","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982695","url":null,"abstract":"We propose a novel approach for detecting keypoints invariant to scale changes based on M-wavelet theory. The theory description and detecting process of our approach are presented The comparative evaluation of different detectors shows our approach can provides a competent performance in rotation invariant, scale invariant, illumination invariant and noiseproof. In terms of scale changes, our proposed approach improves keypoint repeatability by 2%~10% compared with scale invariant feature transform (SIFT), speeded up robust features (SURF), Harris-Laplace, Hessian-Laplace.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"161 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116418208","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982665
Qingyang Li, Yu Zhou, Anlong Ming
Action recognition has a long research history, despite several contributed approaches have been introduced, it remains a challenging task in computer vision. In this paper, we present a uniform fusion framework for action recognition, which integrates not only the local depth cues but also the global depth cues. Firstly, the action recognition task is formulated as the maximize the posterior probability, and then the observation for the original action is decomposed into the sub-observations for each individual feature representation strategy of the original action. For the local depth cues, the joints inside the human skeleton is employed to model the local variation of the human motion. In addition, the normal of the depth surface is utilized as the global cue to capture the holistic structure of the human motion. Rather than using the original feature directly, the support vector machine model learning both the discriminative local cue (i.e., the joint) and the discriminative global cue (i.e., the depth surface), respectively. The presented approach is validated on the famous MSR Daily Activity 3D Dataset. And the experimental results demonstrate that our fusion approach can outperform the baseline approaches.
{"title":"Integrating joint and surface for human action recognition in indoor environments","authors":"Qingyang Li, Yu Zhou, Anlong Ming","doi":"10.1109/SPAC.2014.6982665","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982665","url":null,"abstract":"Action recognition has a long research history, despite several contributed approaches have been introduced, it remains a challenging task in computer vision. In this paper, we present a uniform fusion framework for action recognition, which integrates not only the local depth cues but also the global depth cues. Firstly, the action recognition task is formulated as the maximize the posterior probability, and then the observation for the original action is decomposed into the sub-observations for each individual feature representation strategy of the original action. For the local depth cues, the joints inside the human skeleton is employed to model the local variation of the human motion. In addition, the normal of the depth surface is utilized as the global cue to capture the holistic structure of the human motion. Rather than using the original feature directly, the support vector machine model learning both the discriminative local cue (i.e., the joint) and the discriminative global cue (i.e., the depth surface), respectively. The presented approach is validated on the famous MSR Daily Activity 3D Dataset. And the experimental results demonstrate that our fusion approach can outperform the baseline approaches.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130495405","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982719
Yubei Lin, Qinyi Chen, Xingming Zhang
The latest video coding standard HEVC (High Efficiency Video Coding) provides up to 35 intra prediction modes to achieve more accurate prediction. However, a larger number of prediction modes leads to higher encoding complexity. To accelerate the intra mode decision in HEVC, an improved algorithm based on numerical analysis of the neighboring reference samples of coding units is proposed. Four filters are designed to reduce the candidate modes. And an algorithm to seek the proper thresholds for the filters is developed. The experimental results show that when compared to the test model HM 12.0 of HEVC, the proposed method can save up to 28.40% of encoder time with slight increase of BD-rate and negligible loss of PSNR. Furthermore, it is easy to combine our approach with other fast algorithms to achieve more improvement.
最新的视频编码标准HEVC (High Efficiency video coding)提供多达35种帧内预测模式，以实现更准确的预测。但是，预测模式的数量越多，编码复杂度也就越高。为了加速HEVC中的模式内决策，提出了一种基于相邻编码单元参考样本数值分析的改进算法。设计了四个滤波器来减少候选模式。并提出了一种寻找滤波器合适阈值的算法。实验结果表明，与HEVC的测试模型HM 12.0相比，该方法可节省28.40%的编码器时间，且bd率略有提高，PSNR的损失可以忽略不计。此外，我们的方法很容易与其他快速算法相结合，以实现更多的改进。
{"title":"Numerical analysis based fast intra prediction algorithm in HEVC","authors":"Yubei Lin, Qinyi Chen, Xingming Zhang","doi":"10.1109/SPAC.2014.6982719","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982719","url":null,"abstract":"The latest video coding standard HEVC (High Efficiency Video Coding) provides up to 35 intra prediction modes to achieve more accurate prediction. However, a larger number of prediction modes leads to higher encoding complexity. To accelerate the intra mode decision in HEVC, an improved algorithm based on numerical analysis of the neighboring reference samples of coding units is proposed. Four filters are designed to reduce the candidate modes. And an algorithm to seek the proper thresholds for the filters is developed. The experimental results show that when compared to the test model HM 12.0 of HEVC, the proposed method can save up to 28.40% of encoder time with slight increase of BD-rate and negligible loss of PSNR. Furthermore, it is easy to combine our approach with other fast algorithms to achieve more improvement.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134461894","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982671
Guihua Liao, Jian Zhang
This paper presents an information query system for renting and returning public bicycles based on Android system. The system provides two parts of services: position locating and station query. Position locating is tackled by baidu API entirely, station query includes two sub parts: station marking and information updating. We use the latitude and longitude of the station to mark the station on the map, and query the station around current location. We use the station ID as the open API's parameter to query the station information, then analyze the returned JSON data with station information, and extract the remaining bicycles for renting and remaining seats for returning from the JSON data. At last, we update the station information on the map. This system aims to propose an integrated solution for public bicycle service. Experiments in 3 test cases demonstrate the effect of the proposed system, and indicate that the system can fully meet the requirement for public bicycle service.
{"title":"Information query for public bicycle service based on Andriod","authors":"Guihua Liao, Jian Zhang","doi":"10.1109/SPAC.2014.6982671","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982671","url":null,"abstract":"This paper presents an information query system for renting and returning public bicycles based on Android system. The system provides two parts of services: position locating and station query. Position locating is tackled by baidu API entirely, station query includes two sub parts: station marking and information updating. We use the latitude and longitude of the station to mark the station on the map, and query the station around current location. We use the station ID as the open API's parameter to query the station information, then analyze the returned JSON data with station information, and extract the remaining bicycles for renting and remaining seats for returning from the JSON data. At last, we update the station information on the map. This system aims to propose an integrated solution for public bicycle service. Experiments in 3 test cases demonstrate the effect of the proposed system, and indicate that the system can fully meet the requirement for public bicycle service.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"201 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124516458","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982727
R. Luo, Xuejia Lai, R. You
In this paper, we propose an improved table-based white-box implementation of AES which is able to resist different types of attack, including the BGE attack and De Mulder et al.'s cryptanalysis, to protect information under “white-box attack context”. The notion of white-box attack context, introduced by Chow et al., describes a general setting in which cryptographic algorithms are executed in untrusted environments. In this setting, adversaries have attained complete access to the implementations of cryptographic algorithms as well as the dynamic execution environments. The key strategy applied to our design is to compose different operations of the AES round function and convert the composition into encoded lookup tables. The new scheme exploits larger key-dependent tables, each of which contains two bytes of the round keys. We then analyze the security against different types of attack and measure two security metrics: the “white-box diversity” and “ambiguity”. The new scheme can withstand the BGE attack due to the utilization of larger mixing bijections and tabulated “ShiftRows” it can also resist the cryptanalysis of De Mulder et al. since the bindings between “nTMC” and “TSR” are irreducible and the non-linear encodings are introduced to all tables.
{"title":"A new attempt of white-box AES implementation","authors":"R. Luo, Xuejia Lai, R. You","doi":"10.1109/SPAC.2014.6982727","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982727","url":null,"abstract":"In this paper, we propose an improved table-based white-box implementation of AES which is able to resist different types of attack, including the BGE attack and De Mulder et al.'s cryptanalysis, to protect information under “white-box attack context”. The notion of white-box attack context, introduced by Chow et al., describes a general setting in which cryptographic algorithms are executed in untrusted environments. In this setting, adversaries have attained complete access to the implementations of cryptographic algorithms as well as the dynamic execution environments. The key strategy applied to our design is to compose different operations of the AES round function and convert the composition into encoded lookup tables. The new scheme exploits larger key-dependent tables, each of which contains two bytes of the round keys. We then analyze the security against different types of attack and measure two security metrics: the “white-box diversity” and “ambiguity”. The new scheme can withstand the BGE attack due to the utilization of larger mixing bijections and tabulated “ShiftRows” it can also resist the cryptanalysis of De Mulder et al. since the bindings between “nTMC” and “TSR” are irreducible and the non-linear encodings are introduced to all tables.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124602959","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}
This paper proposes a general solution for the multi-source high resolution real-time monitoring system. Our proposed system is realized on a PCI bus-based real-time image processing board, based on H.264 video compression standards. The main function of the system includes video decompression, image scaling, image stitching, as well as multi-mode real-time display in PC software. Unlike traditional multi-board based multi-source surveillance system, our solution focuses on single board system with synchronously processing and independently display of multi-source video, which is more resource-saving The main contributions of this paper are twofold: firstly, we propose a new logical control method of multi-thread to overcome the limitation of single-board bandwidth, and secondly we give a DirectDraw based display solution to avoid screen flicker caused by different frame rates of multi-source. In addition, we elaborate an approach on multi-thread control and synchronization to reduce CPU utility and monitor abnormal system condition.
{"title":"A general solution for multi-thread based multi-source compressed video surveillance system","authors":"Lu Han, Yixiao Zhao, Shujian Yu, Baojun Zhao, Jiatong Li, Jinghui Wu","doi":"10.1109/SPAC.2014.6982664","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982664","url":null,"abstract":"This paper proposes a general solution for the multi-source high resolution real-time monitoring system. Our proposed system is realized on a PCI bus-based real-time image processing board, based on H.264 video compression standards. The main function of the system includes video decompression, image scaling, image stitching, as well as multi-mode real-time display in PC software. Unlike traditional multi-board based multi-source surveillance system, our solution focuses on single board system with synchronously processing and independently display of multi-source video, which is more resource-saving The main contributions of this paper are twofold: firstly, we propose a new logical control method of multi-thread to overcome the limitation of single-board bandwidth, and secondly we give a DirectDraw based display solution to avoid screen flicker caused by different frame rates of multi-source. In addition, we elaborate an approach on multi-thread control and synchronization to reduce CPU utility and monitor abnormal system condition.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122592607","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982731
Jia Wang, Zhijun Song, Qian Li, Jun Yu, Fei Chen
In order to overcome the limitation of existing data cleansing methods working on massive data, in this paper, we propose a generic semantic-based framework using parallelized processing model for effective big data cleansing. We also use an improved Semantic-Based Keyword Matching Algorithm to deal with duplicate data. Experimental results show that this parallelized framework with improved Semantic-Based Keyword Matching Algorithm can identify duplicates with high recall and precision and have a good performance for big data cleansing.
{"title":"Semantic-based intelligent data clean framework for big data","authors":"Jia Wang, Zhijun Song, Qian Li, Jun Yu, Fei Chen","doi":"10.1109/SPAC.2014.6982731","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982731","url":null,"abstract":"In order to overcome the limitation of existing data cleansing methods working on massive data, in this paper, we propose a generic semantic-based framework using parallelized processing model for effective big data cleansing. We also use an improved Semantic-Based Keyword Matching Algorithm to deal with duplicate data. Experimental results show that this parallelized framework with improved Semantic-Based Keyword Matching Algorithm can identify duplicates with high recall and precision and have a good performance for big data cleansing.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117229988","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}
Pub Date : 2014-12-15DOI: 10.1109/SPAC.2014.6982649
Xiaoyu Hu, Hongquan Yun, Delie Ming, Tian Tian
In this paper, we propose a compressive tracking method via appearance model based on structural local patchs and improved Haar-like feature. In contrast to previous compressive tracking only considering the holistic representation, an object can be represented by local image patches with spatial layout in an object. This representation takes advantage of both partial information and spatial information of the target. Each local patch has a fixed position in the target field, and all local patches can represent the whole target. In addition, our appearance model based on features extracted from image patches, which can guarantee the randomness of the rectangular boxes and the distribution of the rectangular boxes over the entire image area, avoiding the randomness of the rectangular boxes is too strong to weak the feature expression. We sample the positive and negative samples and divide them into patchs to train a binary classification via a naive Bayes classifier with online update, then the classifier is used to discriminate the candidate samples. The candidate sample which gets the highest classify score is the target. After that we draw positive and negative samples in the same way with the candidate samples to update the classifier to get ready for next frame. Our approach helps not only locate the target more accurately but also can handle partial occlusion effectively. The proposed tracker is compared with several state-of-the-art trackers on some challenging video sequences. Our proposed tracker is better and more stable in both quantitative and qualitative comparisons.
{"title":"Compressive tracking via appearance modeling based on structural local patch","authors":"Xiaoyu Hu, Hongquan Yun, Delie Ming, Tian Tian","doi":"10.1109/SPAC.2014.6982649","DOIUrl":"https://doi.org/10.1109/SPAC.2014.6982649","url":null,"abstract":"In this paper, we propose a compressive tracking method via appearance model based on structural local patchs and improved Haar-like feature. In contrast to previous compressive tracking only considering the holistic representation, an object can be represented by local image patches with spatial layout in an object. This representation takes advantage of both partial information and spatial information of the target. Each local patch has a fixed position in the target field, and all local patches can represent the whole target. In addition, our appearance model based on features extracted from image patches, which can guarantee the randomness of the rectangular boxes and the distribution of the rectangular boxes over the entire image area, avoiding the randomness of the rectangular boxes is too strong to weak the feature expression. We sample the positive and negative samples and divide them into patchs to train a binary classification via a naive Bayes classifier with online update, then the classifier is used to discriminate the candidate samples. The candidate sample which gets the highest classify score is the target. After that we draw positive and negative samples in the same way with the candidate samples to update the classifier to get ready for next frame. Our approach helps not only locate the target more accurately but also can handle partial occlusion effectively. The proposed tracker is compared with several state-of-the-art trackers on some challenging video sequences. Our proposed tracker is better and more stable in both quantitative and qualitative comparisons.","PeriodicalId":326246,"journal":{"name":"Proceedings 2014 IEEE International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":"27 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129498386","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：