All existing feature point based partial-duplicate image retrieval systems are confronted with the false feature point matching problem. To resolve this issue, geometric contexts are widely used to verify the geometric consistency in order to remove false matches. However, most of the existing methods focus on local geometric contexts rather than global. Seeking global contexts has attracted a lot of attention in recent years. This paper introduces a novel global geometric consistency, based on the low rankness of squared distance matrices of feature points, to detect false matches. We cast the problem of detecting false matches as a problem of decomposing a squared distance matrix into a low rank matrix, which models the global geometric consistency, and a sparse matrix, which models the mismatched feature points. So we arrive at a model of Robust Principal Component Analysis. Our Low Rank Global Geometric Consistency (LRGGC) is simple yet effective and theoretically sound. Extensive experimental results show that our LRGGC is much more accurate than state of the art geometric verification methods in detecting false matches and is robust to all kinds of similarity transformation (scaling, rotation, and translation) and even slight change in 3D views. Its speed is also highly competitive even compared with local geometric consistency based methods.
{"title":"Low Rank Global Geometric Consistency for Partial-Duplicate Image Search","authors":"Li Yang, Yang Lin, Zhouchen Lin, H. Zha","doi":"10.1109/ICPR.2014.675","DOIUrl":"https://doi.org/10.1109/ICPR.2014.675","url":null,"abstract":"All existing feature point based partial-duplicate image retrieval systems are confronted with the false feature point matching problem. To resolve this issue, geometric contexts are widely used to verify the geometric consistency in order to remove false matches. However, most of the existing methods focus on local geometric contexts rather than global. Seeking global contexts has attracted a lot of attention in recent years. This paper introduces a novel global geometric consistency, based on the low rankness of squared distance matrices of feature points, to detect false matches. We cast the problem of detecting false matches as a problem of decomposing a squared distance matrix into a low rank matrix, which models the global geometric consistency, and a sparse matrix, which models the mismatched feature points. So we arrive at a model of Robust Principal Component Analysis. Our Low Rank Global Geometric Consistency (LRGGC) is simple yet effective and theoretically sound. Extensive experimental results show that our LRGGC is much more accurate than state of the art geometric verification methods in detecting false matches and is robust to all kinds of similarity transformation (scaling, rotation, and translation) and even slight change in 3D views. Its speed is also highly competitive even compared with local geometric consistency based methods.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131940811","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}
Background subtraction is a popular technique used in accurate foreground extraction with a stationary background. Since most outdoor surveillance videos are taken in complex environments, their "stationary" backgrounds change in some unknown patterns, which make the perfect foreground extraction very difficult. Based on visual background extractor (ViBe) scheme, in this paper we propose a new background subtraction algorithm which includes two innovative mechanisms and several other improved technique tricks. The paper inherits and develops background modeling based on pixel sample values, and use dynamic noise sampling and complementary learning to overcome the pixel-wise background model's intrinsic shortcomings. Besides, the algorithm works on the quantitative analysis without any estimation of the probability density function (pdf). Hence, it takes relatively low computational cost. Extensive experiments on a popular public dataset show that the proposed method has much better precision than ViBe, and could get the best precision and the highest average ranking compared with 27 state-of-the-art algorithms presented on the change detection website.
{"title":"Background Subtraction with Dynamic Noise Sampling and Complementary Learning","authors":"Weifeng Ge, Yuhan Dong, Zhenhua Guo, Youbin Chen","doi":"10.1109/ICPR.2014.406","DOIUrl":"https://doi.org/10.1109/ICPR.2014.406","url":null,"abstract":"Background subtraction is a popular technique used in accurate foreground extraction with a stationary background. Since most outdoor surveillance videos are taken in complex environments, their \"stationary\" backgrounds change in some unknown patterns, which make the perfect foreground extraction very difficult. Based on visual background extractor (ViBe) scheme, in this paper we propose a new background subtraction algorithm which includes two innovative mechanisms and several other improved technique tricks. The paper inherits and develops background modeling based on pixel sample values, and use dynamic noise sampling and complementary learning to overcome the pixel-wise background model's intrinsic shortcomings. Besides, the algorithm works on the quantitative analysis without any estimation of the probability density function (pdf). Hence, it takes relatively low computational cost. Extensive experiments on a popular public dataset show that the proposed method has much better precision than ViBe, and could get the best precision and the highest average ranking compared with 27 state-of-the-art algorithms presented on the change detection website.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130967436","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}
Lisa Senger, M. Schröer, J. H. Metzen, E. Kirchner
In order to transfer complex human behavior to a robot, segmentation methods are needed which are able to detect central movement patterns that can be combined to generate a wide range of behaviors. We propose an algorithm that segments human movements into behavior building blocks in a fully automatic way, called velocity-based Multiple Change-point Inference (vMCI). Based on characteristic bell-shaped velocity patterns that can be found in point-to-point arm movements, the algorithm infers segment borders using Bayesian inference. Different segment lengths and variations in the movement execution can be handled. Moreover, the number of segments the movement is composed of need not be known in advance. Several experiments are performed on synthetic and motion capturing data of human movements to compare vMCI with other techniques for unsupervised segmentation. The results show that vMCI is able to detect segment borders even in noisy data and in demonstrations with smooth transitions between segments.
{"title":"Velocity-Based Multiple Change-Point Inference for Unsupervised Segmentation of Human Movement Behavior","authors":"Lisa Senger, M. Schröer, J. H. Metzen, E. Kirchner","doi":"10.1109/ICPR.2014.781","DOIUrl":"https://doi.org/10.1109/ICPR.2014.781","url":null,"abstract":"In order to transfer complex human behavior to a robot, segmentation methods are needed which are able to detect central movement patterns that can be combined to generate a wide range of behaviors. We propose an algorithm that segments human movements into behavior building blocks in a fully automatic way, called velocity-based Multiple Change-point Inference (vMCI). Based on characteristic bell-shaped velocity patterns that can be found in point-to-point arm movements, the algorithm infers segment borders using Bayesian inference. Different segment lengths and variations in the movement execution can be handled. Moreover, the number of segments the movement is composed of need not be known in advance. Several experiments are performed on synthetic and motion capturing data of human movements to compare vMCI with other techniques for unsupervised segmentation. The results show that vMCI is able to detect segment borders even in noisy data and in demonstrations with smooth transitions between segments.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114236715","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}
Based on fitting the Local Binary Patterns (LBP) histogram into the bag-of-words paradigm, we propose an LBP variant termed Principal Local Binary Patterns (PLBP) which are learned in an unsupervised way from the data. The learning problem turns out to be the same as the Principal Component Analysis (PCA) and thus can be solved very efficiently. Unlike the manually specified patterns in LBP which distribute very non-uniformly, the learned patterns in PLBP can adapt with the distribution of the data so that they distribute very uniformly, which preserves more information than LBP in the binary coding process. Moreover, PLBP can take advantage of much larger neighborhood than LBP to describe the point, which provides more information. Therefore, PLBP contains more information than LBP to discriminate different classes. The experimental results of face recognition on the FERET and LFW datasets clearly confirm the discrimination power and robustness of PLBP. It achieves very competing performance on both datasets and it is very simple and efficient to compute.
{"title":"Principal Local Binary Patterns for Face Representation and Recognition","authors":"J. Yi, Fei Su","doi":"10.1109/ICPR.2014.779","DOIUrl":"https://doi.org/10.1109/ICPR.2014.779","url":null,"abstract":"Based on fitting the Local Binary Patterns (LBP) histogram into the bag-of-words paradigm, we propose an LBP variant termed Principal Local Binary Patterns (PLBP) which are learned in an unsupervised way from the data. The learning problem turns out to be the same as the Principal Component Analysis (PCA) and thus can be solved very efficiently. Unlike the manually specified patterns in LBP which distribute very non-uniformly, the learned patterns in PLBP can adapt with the distribution of the data so that they distribute very uniformly, which preserves more information than LBP in the binary coding process. Moreover, PLBP can take advantage of much larger neighborhood than LBP to describe the point, which provides more information. Therefore, PLBP contains more information than LBP to discriminate different classes. The experimental results of face recognition on the FERET and LFW datasets clearly confirm the discrimination power and robustness of PLBP. It achieves very competing performance on both datasets and it is very simple and efficient to compute.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116303776","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 cost-sensitive transformation for improving handwritten address recognition performance by converting a general-purpose handwritten Chinese character recognition engine to a special-purpose one. The class probabilities produced by character recognition engine for predicting a sample to candidate classes are transformed to the expected costs based on Naive Bayes optimal theoretical predictions firstly. And then candidate probabilities are reestimated based on the expected costs. Two general-purpose offline handwritten Chinese character recognition engines, PAIS and HAW, are tested in our experiments by applying them in handwritten Chinese address recognition system. 1822 live handwritten Chinese address images are tested with multiple cost matrices. Experimental results show that cost-sensitive transformation improves the recognition performance of general purpose recognition engines on handwritten Chinese address recognition.
{"title":"Cost-Sensitive Transformation for Chinese Address Recognition","authors":"Shujing Lu, Xiaohua Wei, Yue Lu","doi":"10.1109/ICPR.2014.499","DOIUrl":"https://doi.org/10.1109/ICPR.2014.499","url":null,"abstract":"This paper proposes a cost-sensitive transformation for improving handwritten address recognition performance by converting a general-purpose handwritten Chinese character recognition engine to a special-purpose one. The class probabilities produced by character recognition engine for predicting a sample to candidate classes are transformed to the expected costs based on Naive Bayes optimal theoretical predictions firstly. And then candidate probabilities are reestimated based on the expected costs. Two general-purpose offline handwritten Chinese character recognition engines, PAIS and HAW, are tested in our experiments by applying them in handwritten Chinese address recognition system. 1822 live handwritten Chinese address images are tested with multiple cost matrices. Experimental results show that cost-sensitive transformation improves the recognition performance of general purpose recognition engines on handwritten Chinese address recognition.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122443449","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}
Imposing a spatial coherence constraint on image matching is becoming a necessity for local feature based object retrieval. We tackle the affine invariance problem of the prior spatial coherence model and propose a novel approach for geometrically stable image retrieval. Compared with related studies focusing simply on translation, rotation, and isotropic scaling, our approach can deal with more significant transformations including anisotropic scaling and shearing. Our contribution consists of revisiting the first-order affine adaptation approach and extending its application to represent the geometric coherence of a second-order local feature structure. We comprehensively evaluated our approach using Flickr Logos 32, Holiday, and Oxford Buildings benchmarks. Extensive experimentation and comparisons with state-of-the-art spatial coherence models demonstrate the superiority of our approach in image retrieval tasks.
{"title":"Image Retrieval Based on Anisotropic Scaling and Shearing Invariant Geometric Coherence","authors":"Xiaomeng Wu, K. Kashino","doi":"10.1109/ICPR.2014.677","DOIUrl":"https://doi.org/10.1109/ICPR.2014.677","url":null,"abstract":"Imposing a spatial coherence constraint on image matching is becoming a necessity for local feature based object retrieval. We tackle the affine invariance problem of the prior spatial coherence model and propose a novel approach for geometrically stable image retrieval. Compared with related studies focusing simply on translation, rotation, and isotropic scaling, our approach can deal with more significant transformations including anisotropic scaling and shearing. Our contribution consists of revisiting the first-order affine adaptation approach and extending its application to represent the geometric coherence of a second-order local feature structure. We comprehensively evaluated our approach using Flickr Logos 32, Holiday, and Oxford Buildings benchmarks. Extensive experimentation and comparisons with state-of-the-art spatial coherence models demonstrate the superiority of our approach in image retrieval tasks.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122755991","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}
Segmentation of handwritten text into individual characters is an important step in many handwriting recognition tasks. In this paper, we present two segmentation algorithms based on elastic shape analysis of parameterized, planar curves. The shape analysis methodology provides matching, comparison and averaging of handwritten curves in a unified framework, which are very useful tools for designing segmentation algorithms. The first type of segmentation can be performed by splitting a full word into individual characters using a matching function. Another type of segmentation can be obtained by matching parts of the handwritten words to a given individual character. We validate the two proposed algorithms on real handwritten signatures and words coming from the SVC 2004 and the UNIPEN ICROW 2003 datasets. We show that the proposed methods are able to successfully segment text coming from highly variable handwriting styles.
{"title":"Handwritten Text Segmentation Using Elastic Shape Analysis","authors":"S. Kurtek, Anuj Srivastava","doi":"10.1109/ICPR.2014.432","DOIUrl":"https://doi.org/10.1109/ICPR.2014.432","url":null,"abstract":"Segmentation of handwritten text into individual characters is an important step in many handwriting recognition tasks. In this paper, we present two segmentation algorithms based on elastic shape analysis of parameterized, planar curves. The shape analysis methodology provides matching, comparison and averaging of handwritten curves in a unified framework, which are very useful tools for designing segmentation algorithms. The first type of segmentation can be performed by splitting a full word into individual characters using a matching function. Another type of segmentation can be obtained by matching parts of the handwritten words to a given individual character. We validate the two proposed algorithms on real handwritten signatures and words coming from the SVC 2004 and the UNIPEN ICROW 2003 datasets. We show that the proposed methods are able to successfully segment text coming from highly variable handwriting styles.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122237854","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}
Deformable part models show a remarkable detection performance for a variety of object categories. During training these models rely on energy-based methods and heuristic initialization to search and localize parts, equivalent to learning local object features. Due to weak supervision, however, those learnt part detectors contain lots of noise and are not enough reliable to classify the object. This paper investigates part localization problem and extends the latent-SVM by incorporating local consistency of image features. The objective is to adaptively select part sub-windows that overlap semantically meaningful components as much as possible, which leads to a more reliable learning of the part detectors in a weakly-supervised setting. The main idea of our method is estimating part-specific color/texture models as well as edge distribution within each training example, followed by a foreground segmentation for part localization. The experimental results show that we achieve an overall improvement of about 3% mAP over the latent-SVM on non-rigid objects.
{"title":"Effective Part Localization in Latent-SVM Training","authors":"Yaodong Chen, Renfa Li","doi":"10.1109/ICPR.2014.732","DOIUrl":"https://doi.org/10.1109/ICPR.2014.732","url":null,"abstract":"Deformable part models show a remarkable detection performance for a variety of object categories. During training these models rely on energy-based methods and heuristic initialization to search and localize parts, equivalent to learning local object features. Due to weak supervision, however, those learnt part detectors contain lots of noise and are not enough reliable to classify the object. This paper investigates part localization problem and extends the latent-SVM by incorporating local consistency of image features. The objective is to adaptively select part sub-windows that overlap semantically meaningful components as much as possible, which leads to a more reliable learning of the part detectors in a weakly-supervised setting. The main idea of our method is estimating part-specific color/texture models as well as edge distribution within each training example, followed by a foreground segmentation for part localization. The experimental results show that we achieve an overall improvement of about 3% mAP over the latent-SVM on non-rigid objects.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129140805","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}
Ensemble clustering aggregates partitions obtained from several individual clustering algorithms. This can improve the accuracy of results from individual methods and provide robustness against variability in the methods applied. Theorems show one can find dominant sets (clusters) very efficiently by using an evolutionary game theoretic approach. Experiments on an MRI data set consisting of about 4 million data are detailed. The distributed dominant set framework generates partitions of quality slightly better than clustering all the data using fuzzy C means.
{"title":"Dominant Sets as a Framework for Cluster Ensembles: An Evolutionary Game Theory Approach","authors":"Alireza Chakeri, L. Hall","doi":"10.1109/ICPR.2014.595","DOIUrl":"https://doi.org/10.1109/ICPR.2014.595","url":null,"abstract":"Ensemble clustering aggregates partitions obtained from several individual clustering algorithms. This can improve the accuracy of results from individual methods and provide robustness against variability in the methods applied. Theorems show one can find dominant sets (clusters) very efficiently by using an evolutionary game theoretic approach. Experiments on an MRI data set consisting of about 4 million data are detailed. The distributed dominant set framework generates partitions of quality slightly better than clustering all the data using fuzzy C means.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132243579","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 describes a new hybrid detection method that combines exemplar based approach with discriminative patch selection. More specifically, we applied a modified random forest for retrieval of input similar local patches of stored exemplars while rejecting background patches. A recursive algorithm based on dynamic programming 2D matching optimization is applied after the aforementioned patch retrieving stage in order to enforce geometric constraints of object patches. Our proposed approach demonstrates experimentally that it performs well while maintaining the capability for incremental learning.
{"title":"Compound Exemplar Based Object Detection by Incremental Random Forest","authors":"Kai Ma, J. Ben-Arie","doi":"10.1109/ICPR.2014.417","DOIUrl":"https://doi.org/10.1109/ICPR.2014.417","url":null,"abstract":"This paper describes a new hybrid detection method that combines exemplar based approach with discriminative patch selection. More specifically, we applied a modified random forest for retrieval of input similar local patches of stored exemplars while rejecting background patches. A recursive algorithm based on dynamic programming 2D matching optimization is applied after the aforementioned patch retrieving stage in order to enforce geometric constraints of object patches. Our proposed approach demonstrates experimentally that it performs well while maintaining the capability for incremental learning.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130612985","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：