Low-rank matrix completion has recently emerged in computational data analysis. The problem aims to recover a low-rank representation from the contaminated data. The errors in data are assumed to be sparse, which is generally characterized by minimizing the L1-norm of the residual. This actually assumes that the residual follows the Laplacian distribution. The Laplacian assumption, however, may not be accurate enough to describe various noises in real scenarios. In this paper, we estimate the error in data with robust regression. Assuming the noises are respectively independent and identically distributed, the minimization of noise is equivalent to find the maximum likelihood estimation (MLE) solution for the residuals. We also design an iteratively reweight inexact augmented Lagrange multiplier algorithm to solve the optimization. Experimental results confirm the efficiency of our proposed approach under different conditions.
{"title":"Low-Rank Matrix Completion Based on Maximum Likelihood Estimation","authors":"Jinhui Chen, Jian Yang","doi":"10.1109/ACPR.2013.120","DOIUrl":"https://doi.org/10.1109/ACPR.2013.120","url":null,"abstract":"Low-rank matrix completion has recently emerged in computational data analysis. The problem aims to recover a low-rank representation from the contaminated data. The errors in data are assumed to be sparse, which is generally characterized by minimizing the L1-norm of the residual. This actually assumes that the residual follows the Laplacian distribution. The Laplacian assumption, however, may not be accurate enough to describe various noises in real scenarios. In this paper, we estimate the error in data with robust regression. Assuming the noises are respectively independent and identically distributed, the minimization of noise is equivalent to find the maximum likelihood estimation (MLE) solution for the residuals. We also design an iteratively reweight inexact augmented Lagrange multiplier algorithm to solve the optimization. Experimental results confirm the efficiency of our proposed approach under different conditions.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126796187","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}
Real-world data sets are highly complicated. They can contain a lot of features, and may involve multiple learning tasks with intrinsically or explicitly represented task relationships. In this paper, we briefly discuss several recent approaches that can be used in these scenarios. The algorithms presented are flexible in capturing the task relationships, computationally efficient with good scalability, and have better empirical performance than the existing approaches.
{"title":"Learning from High-Dimensional Data in Multitask/Multilabel Classification","authors":"J. Kwok","doi":"10.1109/ACPR.2013.214","DOIUrl":"https://doi.org/10.1109/ACPR.2013.214","url":null,"abstract":"Real-world data sets are highly complicated. They can contain a lot of features, and may involve multiple learning tasks with intrinsically or explicitly represented task relationships. In this paper, we briefly discuss several recent approaches that can be used in these scenarios. The algorithms presented are flexible in capturing the task relationships, computationally efficient with good scalability, and have better empirical performance than the existing approaches.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121591310","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}
Hiroyuki Ogihara, Y. Fujita, Y. Hamamoto, N. Iizuka, M. Oka
Liver cancer has a high likelihood of recurrence despite complete surgical resection and is thus known as an intractable cancer. If postoperative recurrence of cancer is correctly predicted for each patient as a form of personalized medicine, effective treatment can be carried out. The purpose of this paper is to investigate prediction of recurrence of liver cancer by use of blood test data only in patients who underwent complete surgical resection of liver cancer. For this purpose, we propose a classifier based on Boolean algebra using a binary pattern consisting of a combination of clinical and genomic data by which we can predict recurrence of liver cancer. We perform a predictive experiment using data from patients with recurrence and non-recurrence and discuss the effectiveness of the proposed method from the experimental results.
{"title":"Classification Based on Boolean Algebra and Its Application to the Prediction of Recurrence of Liver Cancer","authors":"Hiroyuki Ogihara, Y. Fujita, Y. Hamamoto, N. Iizuka, M. Oka","doi":"10.1109/ACPR.2013.152","DOIUrl":"https://doi.org/10.1109/ACPR.2013.152","url":null,"abstract":"Liver cancer has a high likelihood of recurrence despite complete surgical resection and is thus known as an intractable cancer. If postoperative recurrence of cancer is correctly predicted for each patient as a form of personalized medicine, effective treatment can be carried out. The purpose of this paper is to investigate prediction of recurrence of liver cancer by use of blood test data only in patients who underwent complete surgical resection of liver cancer. For this purpose, we propose a classifier based on Boolean algebra using a binary pattern consisting of a combination of clinical and genomic data by which we can predict recurrence of liver cancer. We perform a predictive experiment using data from patients with recurrence and non-recurrence and discuss the effectiveness of the proposed method from the experimental results.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114404301","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 method to localize a mobile camera by searching corresponding points between the mobile camera image and aerial images yielded by GIS database. A same object differently appears in the two images, because the mobile camera images are taken from the user's viewpoints (i.e., on the ground), and aerial images from much higher viewpoints. To reduce such differences in appearance, the mobile camera image is transformed into a virtual top-view image by using the gravity information given by an inertia sensor embedded in the mobile camera. The SIFT algorithm is applied to find corresponding points between the virtual top-view and the aerial images. As the result, a homography matrix that transforms the virtual top-view image into the aerial image is obtained. By using the matrix, the position and orientation of mobile camera are estimated. If the textural information about ground region captured by the mobile camera is poor, it is difficult to obtain a sufficient number of correct corresponding points to allow an accurate homography matrix to be calculated. To deal with such cases, we develop an optional process that stitches multiple virtual top-view images together to cover a larger region of the ground. Experimental evaluation is conducted by a developed pilot system.
{"title":"A Mobile Camera Localization Method Using Aerial-View Images","authors":"H. Toriya, I. Kitahara, Y. Ohta","doi":"10.1109/ACPR.2013.27","DOIUrl":"https://doi.org/10.1109/ACPR.2013.27","url":null,"abstract":"This paper proposes a method to localize a mobile camera by searching corresponding points between the mobile camera image and aerial images yielded by GIS database. A same object differently appears in the two images, because the mobile camera images are taken from the user's viewpoints (i.e., on the ground), and aerial images from much higher viewpoints. To reduce such differences in appearance, the mobile camera image is transformed into a virtual top-view image by using the gravity information given by an inertia sensor embedded in the mobile camera. The SIFT algorithm is applied to find corresponding points between the virtual top-view and the aerial images. As the result, a homography matrix that transforms the virtual top-view image into the aerial image is obtained. By using the matrix, the position and orientation of mobile camera are estimated. If the textural information about ground region captured by the mobile camera is poor, it is difficult to obtain a sufficient number of correct corresponding points to allow an accurate homography matrix to be calculated. To deal with such cases, we develop an optional process that stitches multiple virtual top-view images together to cover a larger region of the ground. Experimental evaluation is conducted by a developed pilot system.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116894229","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}
Kazuma Tanaka, M. Iwata, K. Kunze, M. Iwamura, K. Kise
In this paper we describe a novel annotation service which is capable of seamlessly linking physical and digital worlds through paper documents. Our service uses a real-time document image retrieval method called Locally Likely Arrangement Hashing (LLAH) for providing information associated with the document. By using this service digital annotations can be added to physical documents and shared with friends via mobile devices. We present the prototype implementation, and provide a discussion covering the technical details of the system.
{"title":"Share Me - A Digital Annotation Sharing Service for Paper Documents with Multiple Clients Support","authors":"Kazuma Tanaka, M. Iwata, K. Kunze, M. Iwamura, K. Kise","doi":"10.1109/ACPR.2013.182","DOIUrl":"https://doi.org/10.1109/ACPR.2013.182","url":null,"abstract":"In this paper we describe a novel annotation service which is capable of seamlessly linking physical and digital worlds through paper documents. Our service uses a real-time document image retrieval method called Locally Likely Arrangement Hashing (LLAH) for providing information associated with the document. By using this service digital annotations can be added to physical documents and shared with friends via mobile devices. We present the prototype implementation, and provide a discussion covering the technical details of the system.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115262678","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 efficacy and efficiency of SIFT have made it a state-of-art feature descriptor. It has been widely used in many computer vision applications such as image classification. A large number of methods, e.g. PCA-SIFT, have been contributed to further improve its performance focusing on different components of it. Differing from those previous works, we broach a new scheme to improve the performance of SIFT's descriptor in this paper. We first establish the connection between SIFT and bag of features (BOF) model in descriptor construction. Based on this connection, we then introduce approaches of BOF, e.g. the preservation of spatial information (we adopt spatial pyramid matching as an example to achieve this goal), into SIFT to enhance its robustness. Experimental results in scene matching and image classification show that the BOF-driven SIFT effectively and consistently outperforms the original SIFT.
{"title":"Improvements to the Descriptor of SIFT by BOF Approaches","authors":"Zhouxin Yang, Takio Kurita","doi":"10.1109/ACPR.2013.31","DOIUrl":"https://doi.org/10.1109/ACPR.2013.31","url":null,"abstract":"The efficacy and efficiency of SIFT have made it a state-of-art feature descriptor. It has been widely used in many computer vision applications such as image classification. A large number of methods, e.g. PCA-SIFT, have been contributed to further improve its performance focusing on different components of it. Differing from those previous works, we broach a new scheme to improve the performance of SIFT's descriptor in this paper. We first establish the connection between SIFT and bag of features (BOF) model in descriptor construction. Based on this connection, we then introduce approaches of BOF, e.g. the preservation of spatial information (we adopt spatial pyramid matching as an example to achieve this goal), into SIFT to enhance its robustness. Experimental results in scene matching and image classification show that the BOF-driven SIFT effectively and consistently outperforms the original SIFT.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115672950","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}
Sparse representations have emerged as a powerful approach for encoding images in a large class of machine recognition problems including face recognition. These methods rely on the use of an over-complete basis set for representing an image. This often assumes the availability of a large number of labeled training images, especially for high dimensional data. In many practical problems, the number of labeled training samples are very limited leading to significant degradations in classification performance. To address the problem of lack of training samples, we propose a semi-supervised algorithm that labels the unlabeled samples through a multi-stage label propagation combined with sparse representation. In this representation, each image is decomposed as a linear combination of its nearest basis images, which has the advantage of both locality and sparsity. Extensive experiments on publicly available face databases show that the results are significantly better compared to state-of-the-art face recognition methods in semi-supervised setting and are on par with fully supervised techniques.
{"title":"Sparse Representation Based Face Recognition with Limited Labeled Samples","authors":"Vijay Kumar, A. Namboodiri, C. V. Jawahar","doi":"10.1109/ACPR.2013.38","DOIUrl":"https://doi.org/10.1109/ACPR.2013.38","url":null,"abstract":"Sparse representations have emerged as a powerful approach for encoding images in a large class of machine recognition problems including face recognition. These methods rely on the use of an over-complete basis set for representing an image. This often assumes the availability of a large number of labeled training images, especially for high dimensional data. In many practical problems, the number of labeled training samples are very limited leading to significant degradations in classification performance. To address the problem of lack of training samples, we propose a semi-supervised algorithm that labels the unlabeled samples through a multi-stage label propagation combined with sparse representation. In this representation, each image is decomposed as a linear combination of its nearest basis images, which has the advantage of both locality and sparsity. Extensive experiments on publicly available face databases show that the results are significantly better compared to state-of-the-art face recognition methods in semi-supervised setting and are on par with fully supervised techniques.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129653543","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}
How do people see a scene? To what do they pay attention in their field of view, and when? This depends on the observer's knowledge, experience, and so on. This study compares the eye movements of an expert and novices, and extracts the skill-based differences in their gaze behaviors. In this paper, we focus on the gaze behaviors of a soccer coach and nonprofessional people while watching a video of a soccer game, and analyze the relationships between the eye movements and dynamic salient objects, that is, the ball and the players, in the video. The results show that, when the ball and some players are near either of the goals, the expert pays attention not to them but to the many other players in the middle of the soccer field. The findings of this study will constitute novel stepping stones for modeling a skillful viewing technique and useful knowledge that can be taught to novices.
{"title":"Analysis of Soccer Coach's Eye Gaze Behavior","authors":"Atsushi Iwatsuki, Takatsugu Hirayama, K. Mase","doi":"10.1109/ACPR.2013.185","DOIUrl":"https://doi.org/10.1109/ACPR.2013.185","url":null,"abstract":"How do people see a scene? To what do they pay attention in their field of view, and when? This depends on the observer's knowledge, experience, and so on. This study compares the eye movements of an expert and novices, and extracts the skill-based differences in their gaze behaviors. In this paper, we focus on the gaze behaviors of a soccer coach and nonprofessional people while watching a video of a soccer game, and analyze the relationships between the eye movements and dynamic salient objects, that is, the ball and the players, in the video. The results show that, when the ball and some players are near either of the goals, the expert pays attention not to them but to the many other players in the middle of the soccer field. The findings of this study will constitute novel stepping stones for modeling a skillful viewing technique and useful knowledge that can be taught to novices.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130128156","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}
Yuta Kamihira, W. Ohyama, T. Wakabayashi, F. Kimura
This paper proposes a new signature verification technique called combined segmentation-verification based on off-line features and on-line features. We use three different off-line feature vectors extracted from full name Japanese signature image and from the sub-images of the first name and the last name. The Mahalanobis distance for each offline feature vector is calculated for signature verification. The on-line feature based technique employs dynamic programming (DP) matching technique for time series data of the signatures. The final decision (verification) is performed by SVM based on the three Mahalanobis distances and the dissimilarity of the DP matching. In the evaluation test the proposed technique achieved 97.22% verification accuracy with even FRR and FAR, which is 3.95% higher than the best accuracy obtained by the individual technique. This result shows that the proposed combined segmentation verification approach improves Japanese signature verification accuracy significantly.
{"title":"Improvement of Japanese Signature Verification by Combined Segmentation Verification Approach","authors":"Yuta Kamihira, W. Ohyama, T. Wakabayashi, F. Kimura","doi":"10.1109/ACPR.2013.46","DOIUrl":"https://doi.org/10.1109/ACPR.2013.46","url":null,"abstract":"This paper proposes a new signature verification technique called combined segmentation-verification based on off-line features and on-line features. We use three different off-line feature vectors extracted from full name Japanese signature image and from the sub-images of the first name and the last name. The Mahalanobis distance for each offline feature vector is calculated for signature verification. The on-line feature based technique employs dynamic programming (DP) matching technique for time series data of the signatures. The final decision (verification) is performed by SVM based on the three Mahalanobis distances and the dissimilarity of the DP matching. In the evaluation test the proposed technique achieved 97.22% verification accuracy with even FRR and FAR, which is 3.95% higher than the best accuracy obtained by the individual technique. This result shows that the proposed combined segmentation verification approach improves Japanese signature verification accuracy significantly.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"84 Pt 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129038212","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}
Wanru Xu, Z. Miao, Jian Zhang, Qiang Zhang, Haohao Wu
In this paper, we propose a new method for human action analysis in videos. A video sequence of human action in our perspective can be modeled through feature distribution over spatial-temporal domain. Relationships between features and each defined action are also explored to form discriminative feature sets. In our work, we first capture contextual correlations between the local features through multiple windows. We then mine confidences from association rules and learn contributions from trained-SVM based on sample videos. Finally, through the analysis of feature distribution and their interactions over spatial-temporal domain, we combine the contexture correlations and the relationships between words and their related actions to derive weights of bag of feature words for action matching. In most of the case, our experiments have indicated that the new method outperforms other previous published results on the Weizmann and KTH datasets.
{"title":"Spatial-Temporal Context for Action Recognition Combined with Confidence and Contribution Weight","authors":"Wanru Xu, Z. Miao, Jian Zhang, Qiang Zhang, Haohao Wu","doi":"10.1109/ACPR.2013.114","DOIUrl":"https://doi.org/10.1109/ACPR.2013.114","url":null,"abstract":"In this paper, we propose a new method for human action analysis in videos. A video sequence of human action in our perspective can be modeled through feature distribution over spatial-temporal domain. Relationships between features and each defined action are also explored to form discriminative feature sets. In our work, we first capture contextual correlations between the local features through multiple windows. We then mine confidences from association rules and learn contributions from trained-SVM based on sample videos. Finally, through the analysis of feature distribution and their interactions over spatial-temporal domain, we combine the contexture correlations and the relationships between words and their related actions to derive weights of bag of feature words for action matching. In most of the case, our experiments have indicated that the new method outperforms other previous published results on the Weizmann and KTH datasets.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127804108","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：