Changsheng Li, L. Yang, Qingshan Liu, F. Meng, Weishan Dong, Yu Wang, Jingmin Xu
In this paper, we propose a new method to learn the regression coefficient matrix for multiple-output regression, which is inspired by multi-task learning. We attempt to incorporate high-order structure information among the regression coefficients into the estimated process of regression coefficient matrix, which is of great importance for multiple-output regression. Meanwhile, we also intend to describe the output structure with noise covariance matrix to assist in learning model parameters. Taking account of the real-world data often corrupted by noise, we place a constraint of minimizing norm on regression coefficient matrix to make it robust to noise. The experiments are conducted on three public available datasets, and the experimental results demonstrate the power of the proposed method against the state-of-the-art methods.
{"title":"Multiple-Output Regression with High-Order Structure Information","authors":"Changsheng Li, L. Yang, Qingshan Liu, F. Meng, Weishan Dong, Yu Wang, Jingmin Xu","doi":"10.1109/ICPR.2014.664","DOIUrl":"https://doi.org/10.1109/ICPR.2014.664","url":null,"abstract":"In this paper, we propose a new method to learn the regression coefficient matrix for multiple-output regression, which is inspired by multi-task learning. We attempt to incorporate high-order structure information among the regression coefficients into the estimated process of regression coefficient matrix, which is of great importance for multiple-output regression. Meanwhile, we also intend to describe the output structure with noise covariance matrix to assist in learning model parameters. Taking account of the real-world data often corrupted by noise, we place a constraint of minimizing norm on regression coefficient matrix to make it robust to noise. The experiments are conducted on three public available datasets, and the experimental results demonstrate the power of the proposed method against the state-of-the-art methods.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"79 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":"127616135","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}
Volkmar Frinken, Yutaro Iwakiri, R. Ishida, Kensho Fujisaki, S. Uchida
At the current rate of technological advancement and social acceptance thereof, it will not be long before wearable devices will be common that constantly record the field of view of the user. We introduce a new database of image sequences, taken with a first person view camera, of realistic, everyday scenes. As a distinguishing feature, we manually transcribed the scene text of each image. This way, sophisticated OCR algorithms can be simulated that can help in the recognition of the location and the activity. To test this hypothesis, we performed a set of experiments using visual features, textual features, and a combination of both. We demonstrate that, although not very powerful when considered alone, the textual information improves the overall recognition rates.
{"title":"Improving Point of View Scene Recognition by Considering Textual Data","authors":"Volkmar Frinken, Yutaro Iwakiri, R. Ishida, Kensho Fujisaki, S. Uchida","doi":"10.1109/ICPR.2014.512","DOIUrl":"https://doi.org/10.1109/ICPR.2014.512","url":null,"abstract":"At the current rate of technological advancement and social acceptance thereof, it will not be long before wearable devices will be common that constantly record the field of view of the user. We introduce a new database of image sequences, taken with a first person view camera, of realistic, everyday scenes. As a distinguishing feature, we manually transcribed the scene text of each image. This way, sophisticated OCR algorithms can be simulated that can help in the recognition of the location and the activity. To test this hypothesis, we performed a set of experiments using visual features, textual features, and a combination of both. We demonstrate that, although not very powerful when considered alone, the textual information improves the overall recognition rates.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132203227","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}
Semi-supervised learning is an important and active topic of research in pattern recognition. For classification using linear discriminant analysis specifically, several semi-supervised variants have been proposed. Using any one of these methods is not guaranteed to outperform the supervised classifier which does not take the additional unlabeled data into account. In this work we compare traditional Expectation Maximization type approaches for semi-supervised linear discriminant analysis with approaches based on intrinsic constraints and propose a new principled approach for semi-supervised linear discriminant analysis, using so-called implicit constraints. We explore the relationships between these methods and consider the question if and in what sense we can expect improvement in performance over the supervised procedure. The constraint based approaches are more robust to misspecification of the model, and may outperform alternatives that make more assumptions on the data in terms of the log-likelihood of unseen objects.
{"title":"Implicitly Constrained Semi-supervised Linear Discriminant Analysis","authors":"J. Krijthe, M. Loog","doi":"10.1109/ICPR.2014.646","DOIUrl":"https://doi.org/10.1109/ICPR.2014.646","url":null,"abstract":"Semi-supervised learning is an important and active topic of research in pattern recognition. For classification using linear discriminant analysis specifically, several semi-supervised variants have been proposed. Using any one of these methods is not guaranteed to outperform the supervised classifier which does not take the additional unlabeled data into account. In this work we compare traditional Expectation Maximization type approaches for semi-supervised linear discriminant analysis with approaches based on intrinsic constraints and propose a new principled approach for semi-supervised linear discriminant analysis, using so-called implicit constraints. We explore the relationships between these methods and consider the question if and in what sense we can expect improvement in performance over the supervised procedure. The constraint based approaches are more robust to misspecification of the model, and may outperform alternatives that make more assumptions on the data in terms of the log-likelihood of unseen objects.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125900396","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}
Nicholas McCarthy, P. Cunningham, Gillian O'Hurley
In this paper we present work on the development of a system for automated classification of digitized H&E histopathology images of prostate carcinoma (PCa). In our system, images are transformed into a tiled grid from which various texture and morphological features are extracted. We evaluate the contribution of high-level morphological features such as those derived from tissue segmentation algorithms as they relate to the accuracy of our classifier models. We also present work on an algorithm for tissue segmentation in image tiles, and introduce a novel feature vector representation of tissue classes in same. Finally, we present the classification accuracy, sensitivity and specificity results of our system when performing three tasks: distinguishing between cancer and non-cancer tiles, between low and high-grade cancer and between Gleason grades 3, 4 and 5. Our results show that the novel tissue representation outperforms the morphological features derived from tissue segmentation by a significant margin, but that neither feature sets improve on the accuracy gained by features from low-level texture methods.
{"title":"The Contribution of Morphological Features in the Classification of Prostate Carcinoma in Digital Pathology Images","authors":"Nicholas McCarthy, P. Cunningham, Gillian O'Hurley","doi":"10.1109/ICPR.2014.563","DOIUrl":"https://doi.org/10.1109/ICPR.2014.563","url":null,"abstract":"In this paper we present work on the development of a system for automated classification of digitized H&E histopathology images of prostate carcinoma (PCa). In our system, images are transformed into a tiled grid from which various texture and morphological features are extracted. We evaluate the contribution of high-level morphological features such as those derived from tissue segmentation algorithms as they relate to the accuracy of our classifier models. We also present work on an algorithm for tissue segmentation in image tiles, and introduce a novel feature vector representation of tissue classes in same. Finally, we present the classification accuracy, sensitivity and specificity results of our system when performing three tasks: distinguishing between cancer and non-cancer tiles, between low and high-grade cancer and between Gleason grades 3, 4 and 5. Our results show that the novel tissue representation outperforms the morphological features derived from tissue segmentation by a significant margin, but that neither feature sets improve on the accuracy gained by features from low-level texture methods.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125429244","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}
Much progress has been made recently in the development of 3D acquisition technologies, which increased the availability of low-cost 3D sensors, such as the Microsoft Kinect. This promotes a wide variety of computer vision applications needing object recognition and 3D reconstruction. We present a novel algorithm for full 3D reconstruction of unknown rotating objects in 2.5D point cloud sequences, such as those generated by 3D sensors. Our algorithm incorporates structural and temporal motion information to build 3D models of moving objects and is based on motion compensated temporal accumulation. The proposed algorithm requires only the fixed centre or axis of rotation, unlike other 3D reconstruction methods, it does not require key point detection, feature description, correspondence matching, provided object models or any geometric information about the object. Moreover, our algorithm integrally estimates the best rigid transformation parameters for registration, applies surface resembling, reduces noise and estimates the optimum angular velocity of the rotating object.
{"title":"3D Moving Object Reconstruction by Temporal Accumulation","authors":"Anas Abuzaina, M. Nixon, J. Carter","doi":"10.1109/ICPR.2014.370","DOIUrl":"https://doi.org/10.1109/ICPR.2014.370","url":null,"abstract":"Much progress has been made recently in the development of 3D acquisition technologies, which increased the availability of low-cost 3D sensors, such as the Microsoft Kinect. This promotes a wide variety of computer vision applications needing object recognition and 3D reconstruction. We present a novel algorithm for full 3D reconstruction of unknown rotating objects in 2.5D point cloud sequences, such as those generated by 3D sensors. Our algorithm incorporates structural and temporal motion information to build 3D models of moving objects and is based on motion compensated temporal accumulation. The proposed algorithm requires only the fixed centre or axis of rotation, unlike other 3D reconstruction methods, it does not require key point detection, feature description, correspondence matching, provided object models or any geometric information about the object. Moreover, our algorithm integrally estimates the best rigid transformation parameters for registration, applies surface resembling, reduces noise and estimates the optimum angular velocity of the rotating object.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"85 22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126300123","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}
S. Tulyakov, R. Vieriu, Stanislau Semeniuta, N. Sebe
This paper proposes a new framework for head pose estimation under extreme pose variations. By augmenting the precision of a template matching based tracking module with the ability to recover offered by a frame-by-frame head pose estimator, we are able to address pose ranges for which face features are no longer visible, while maintaining state-of-the-art performance. Experimental results obtained on a newly acquired 3D extreme head pose dataset support the proposed method and open new perspectives in approaching real-life unconstrained scenarios.
{"title":"Robust Real-Time Extreme Head Pose Estimation","authors":"S. Tulyakov, R. Vieriu, Stanislau Semeniuta, N. Sebe","doi":"10.1109/ICPR.2014.393","DOIUrl":"https://doi.org/10.1109/ICPR.2014.393","url":null,"abstract":"This paper proposes a new framework for head pose estimation under extreme pose variations. By augmenting the precision of a template matching based tracking module with the ability to recover offered by a frame-by-frame head pose estimator, we are able to address pose ranges for which face features are no longer visible, while maintaining state-of-the-art performance. Experimental results obtained on a newly acquired 3D extreme head pose dataset support the proposed method and open new perspectives in approaching real-life unconstrained scenarios.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115506213","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}
The re-use of available labeled samples to classify newly acquired data is a hot topic in pattern analysis and machine learning. Classification algorithms developed with data from one domain cannot be directly used in another related domain, unless the data representation or the classifier have been adapted to the new data distribution. This is crucial in satellite/airborne image analysis: when confronted to domain shifts issued from changes in acquisition or illumination conditions, image classifiers tend to become inaccurate. In this paper, we introduce a method to align data manifolds that represent the same land cover classes, but have undergone spectral distortions. The proposed method relies on a semi-supervised manifold alignment technique and relaxes the requirement of labeled data in all domains by exploiting centrality measures over graphs to match the manifolds. Experiments on multispectral pixel classification at very high spatial resolution show the potential of the method.
{"title":"Unsupervised Alignment of Image Manifolds with Centrality Measures","authors":"D. Tuia, M. Volpi, Gustau Camps-Valls","doi":"10.1109/ICPR.2014.167","DOIUrl":"https://doi.org/10.1109/ICPR.2014.167","url":null,"abstract":"The re-use of available labeled samples to classify newly acquired data is a hot topic in pattern analysis and machine learning. Classification algorithms developed with data from one domain cannot be directly used in another related domain, unless the data representation or the classifier have been adapted to the new data distribution. This is crucial in satellite/airborne image analysis: when confronted to domain shifts issued from changes in acquisition or illumination conditions, image classifiers tend to become inaccurate. In this paper, we introduce a method to align data manifolds that represent the same land cover classes, but have undergone spectral distortions. The proposed method relies on a semi-supervised manifold alignment technique and relaxes the requirement of labeled data in all domains by exploiting centrality measures over graphs to match the manifolds. Experiments on multispectral pixel classification at very high spatial resolution show the potential of the method.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116651165","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}
In this paper, we propose an unsupervised clustering method for axially symmetric directional unit vectors. Our method exploits the Watson distribution and Bregman Divergence within a Model Based Clustering framework. The main objectives of our method are: (a) provide efficient solution to estimate the parameters of a Watson Mixture Model (WMM), (b) generate a set of WMMs and (b) select the optimal model. To this aim, we develop: (a) an efficient soft clustering method, (b) a hierarchical clustering approach in parameter space and (c) a model selection strategy by exploiting information criteria and an evaluation graph. We empirically validate the proposed method using synthetic data. Next, we apply the method for clustering image normals and demonstrate that the proposed method is a potential tool for analyzing the depth image.
{"title":"Unsupervised Clustering of Depth Images Using Watson Mixture Model","authors":"A. Hasnat, O. Alata, A. Trémeau","doi":"10.1109/ICPR.2014.46","DOIUrl":"https://doi.org/10.1109/ICPR.2014.46","url":null,"abstract":"In this paper, we propose an unsupervised clustering method for axially symmetric directional unit vectors. Our method exploits the Watson distribution and Bregman Divergence within a Model Based Clustering framework. The main objectives of our method are: (a) provide efficient solution to estimate the parameters of a Watson Mixture Model (WMM), (b) generate a set of WMMs and (b) select the optimal model. To this aim, we develop: (a) an efficient soft clustering method, (b) a hierarchical clustering approach in parameter space and (c) a model selection strategy by exploiting information criteria and an evaluation graph. We empirically validate the proposed method using synthetic data. Next, we apply the method for clustering image normals and demonstrate that the proposed method is a potential tool for analyzing the depth image.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116769129","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}
Research in automated human gait recognition has largely focused on developing robust feature representation and matching algorithms. In this paper, we investigate the possibility of clustering gait patterns based on the features extracted by automated gait matchers. In this regard, a k-means based clustering approach is used to categorize the feature sets extracted by three different gait matchers. Experiments are conducted in order to determine if (a) the clusters of identities corresponding to the three matchers are similar, and (b) if there is a correlation between gait patterns within each cluster and physical attributes such as gender, body area, height, stride, and cadence. Results demonstrate that human gait patterns can be clustered, where each cluster is defined by identities sharing similar physical attributes. In particular, body area and gender are found to be the primary attributes captured by gait matchers to assess similarity between gait patterns. However, the strength of the correlation between clusters and physical attributes is different across the three matchers, suggesting that gait matchers "weight" attributes differently. The results of this study should be of interest to gait recognition and identification-at-a-distance researchers.
{"title":"On Clustering Human Gait Patterns","authors":"Brian DeCann, A. Ross, M. Culp","doi":"10.1109/ICPR.2014.315","DOIUrl":"https://doi.org/10.1109/ICPR.2014.315","url":null,"abstract":"Research in automated human gait recognition has largely focused on developing robust feature representation and matching algorithms. In this paper, we investigate the possibility of clustering gait patterns based on the features extracted by automated gait matchers. In this regard, a k-means based clustering approach is used to categorize the feature sets extracted by three different gait matchers. Experiments are conducted in order to determine if (a) the clusters of identities corresponding to the three matchers are similar, and (b) if there is a correlation between gait patterns within each cluster and physical attributes such as gender, body area, height, stride, and cadence. Results demonstrate that human gait patterns can be clustered, where each cluster is defined by identities sharing similar physical attributes. In particular, body area and gender are found to be the primary attributes captured by gait matchers to assess similarity between gait patterns. However, the strength of the correlation between clusters and physical attributes is different across the three matchers, suggesting that gait matchers \"weight\" attributes differently. The results of this study should be of interest to gait recognition and identification-at-a-distance researchers.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"260 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120972542","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}
Popular CAPTCHA systems consist of garbled printed text character images with significant distortions and noise. It is believed that humans have little difficulty in deciphering the text, whereas automated systems are foiled by the added noise and distortion. However, in recent years, several text based CAPTCHAs have been reported as broken, that is, automated systems can identify the text in the displayed image with a reasonable amount of success. An extension to the text based CAPTCHA concept is to utilize unconstrained handwritten text, which is still considered to be a challenging problem for automated systems. In this work, we present a automated handwritten CAPTCHA generation system by adding distortions to the Sigma-Lognormal representation of a handwritten word sample. In addition, several noise models are also considered. We perform experiments on the UNIPEN dataset and demonstrate the efficacy of the approach.
{"title":"A Sigma-Lognormal Model for Handwritten Text CAPTCHA Generation","authors":"Chetan Ramaiah, R. Plamondon, V. Govindaraju","doi":"10.1109/ICPR.2014.52","DOIUrl":"https://doi.org/10.1109/ICPR.2014.52","url":null,"abstract":"Popular CAPTCHA systems consist of garbled printed text character images with significant distortions and noise. It is believed that humans have little difficulty in deciphering the text, whereas automated systems are foiled by the added noise and distortion. However, in recent years, several text based CAPTCHAs have been reported as broken, that is, automated systems can identify the text in the displayed image with a reasonable amount of success. An extension to the text based CAPTCHA concept is to utilize unconstrained handwritten text, which is still considered to be a challenging problem for automated systems. In this work, we present a automated handwritten CAPTCHA generation system by adding distortions to the Sigma-Lognormal representation of a handwritten word sample. In addition, several noise models are also considered. We perform experiments on the UNIPEN dataset and demonstrate the efficacy of the approach.","PeriodicalId":142159,"journal":{"name":"2014 22nd International Conference on Pattern Recognition","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121015206","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：