Pub Date : 2002-12-10DOI: 10.1109/ICPR.2002.1048237
H. Moghaddam, Khosrow Amiri Zadeh
In this paper, a new adaptive algorithm for Linear Discriminant Analysis (LDA) is presented. The major advantage of the algorithm is the fast convergence rate, which distinguishes it from the existing on-line methods. Current adaptive methods based on the gradient descent optimization technique use a fixed or a monotonically decreasing step size in each iteration. In this work, we use the steepest descent optimization method to optimally determine the step size in each iteration. It is shown that an optimally variable step size, significantly improves the convergence rate of the algorithm, compared to the conventional methods. The new algorithm has been implemented using a self-organized neural network and its advantages in on-line pattern recognition applications are demonstrated.
{"title":"Fast linear discriminant analysis for on-line pattern recognition applications","authors":"H. Moghaddam, Khosrow Amiri Zadeh","doi":"10.1109/ICPR.2002.1048237","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048237","url":null,"abstract":"In this paper, a new adaptive algorithm for Linear Discriminant Analysis (LDA) is presented. The major advantage of the algorithm is the fast convergence rate, which distinguishes it from the existing on-line methods. Current adaptive methods based on the gradient descent optimization technique use a fixed or a monotonically decreasing step size in each iteration. In this work, we use the steepest descent optimization method to optimally determine the step size in each iteration. It is shown that an optimally variable step size, significantly improves the convergence rate of the algorithm, compared to the conventional methods. The new algorithm has been implemented using a self-organized neural network and its advantages in on-line pattern recognition applications are demonstrated.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126552799","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048449
Hung-Xin Zhao, Yea-Shuan Huang
Tracking human faces is an indispensable process in security access control and automatic video surveillance systems. In this paper, we propose a real-time multiple-person tracking system. Firstly, both skin colour and motion components are extracted to generate face-like regions. Those regions regarded as face candidates are further processed by a silhouette analyser to redefine their boundary and verified by our rule-based verification algorithm. Finally a simple method which utilizes the relationship of skin colour and face history is applied to track multiple persons. Experimental results demonstrate that both the accuracy and processing speed are very promising and can be applied for practical use.
{"title":"Real-time multiple-person tracking system","authors":"Hung-Xin Zhao, Yea-Shuan Huang","doi":"10.1109/ICPR.2002.1048449","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048449","url":null,"abstract":"Tracking human faces is an indispensable process in security access control and automatic video surveillance systems. In this paper, we propose a real-time multiple-person tracking system. Firstly, both skin colour and motion components are extracted to generate face-like regions. Those regions regarded as face candidates are further processed by a silhouette analyser to redefine their boundary and verified by our rule-based verification algorithm. Finally a simple method which utilizes the relationship of skin colour and face history is applied to track multiple persons. Experimental results demonstrate that both the accuracy and processing speed are very promising and can be applied for practical use.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126275988","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048397
J. M. Buenaposada, L. Baumela
In this paper we present a method to estimate in real-time the position and orientation of a previously viewed planar patch. The algorithm is based on minimising the sum of squared differences between a previously stored image of the patch and the current image of it. First a linear model for projectively tracking a planar patch is introduced, then a method to compute the 3D position and orientation of the patch in 3D space is presented. In the experiments conducted we show that this method is adequate for tracking not only planar objects, but also non planar objects with limited out-of-plane rotations, as is the case of face tracking.
{"title":"Real-time tracking and estimation of plane pose","authors":"J. M. Buenaposada, L. Baumela","doi":"10.1109/ICPR.2002.1048397","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048397","url":null,"abstract":"In this paper we present a method to estimate in real-time the position and orientation of a previously viewed planar patch. The algorithm is based on minimising the sum of squared differences between a previously stored image of the patch and the current image of it. First a linear model for projectively tracking a planar patch is introduced, then a method to compute the 3D position and orientation of the patch in 3D space is presented. In the experiments conducted we show that this method is adequate for tracking not only planar objects, but also non planar objects with limited out-of-plane rotations, as is the case of face tracking.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"127 41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114137743","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048400
A. Iwata, K. Kato, Kazuhiko Yamamoto
In this paper, we propose a new camera system called Horizon View Camera (HVC). The HVC is possible to install in the small size robot because the height of the HVC can be short. The HVC is a system in which the optical axis of a camera is directed at the horizon with a mirror so that obtained image contains objects without including the ground itself. Therefore, by using the HVC system, separating objects from the ground becomes very easy. Moreover, there are many other useful features of the HVC system. In order to improve the processing speed and accuracy, we propose a new idea whereby the detection of objects becomes easier and the results are more accurate. Then the HVC serves many uses as the robot vision.
{"title":"The proposal of a new robot vision system called the horizon view camera","authors":"A. Iwata, K. Kato, Kazuhiko Yamamoto","doi":"10.1109/ICPR.2002.1048400","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048400","url":null,"abstract":"In this paper, we propose a new camera system called Horizon View Camera (HVC). The HVC is possible to install in the small size robot because the height of the HVC can be short. The HVC is a system in which the optical axis of a camera is directed at the horizon with a mirror so that obtained image contains objects without including the ground itself. Therefore, by using the HVC system, separating objects from the ground becomes very easy. Moreover, there are many other useful features of the HVC system. In order to improve the processing speed and accuracy, we propose a new idea whereby the detection of objects becomes easier and the results are more accurate. Then the HVC serves many uses as the robot vision.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114211135","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048310
Haiyuan Wu, Genki Yoshikawa, T. Shioyama, S. Lao, M. Kawade
This paper describes a method to detect glasses frames for robust facial image processing. This method makes use of the 3D features obtained by a trinocular stereo vision system. The glasses frame detection is based on the fact that the rims of a pair of glasses lie on the same plane in 3D space. We use a 3D Hough transform to obtain a plane in which 3D features are concentrated. Then, based on the obtained 3D plane and with some geometry constraints, we can detect a group of 3D features belonging to the frame of the glasses. Using this approach, we can separate the 3D features of the glasses frame from those of facial features. This approach does not require any prior knowledge about face pose, eye positions, or the shape of the glasses.
{"title":"Glasses frame detection with 3D Hough transform","authors":"Haiyuan Wu, Genki Yoshikawa, T. Shioyama, S. Lao, M. Kawade","doi":"10.1109/ICPR.2002.1048310","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048310","url":null,"abstract":"This paper describes a method to detect glasses frames for robust facial image processing. This method makes use of the 3D features obtained by a trinocular stereo vision system. The glasses frame detection is based on the fact that the rims of a pair of glasses lie on the same plane in 3D space. We use a 3D Hough transform to obtain a plane in which 3D features are concentrated. Then, based on the obtained 3D plane and with some geometry constraints, we can detect a group of 3D features belonging to the frame of the glasses. Using this approach, we can separate the 3D features of the glasses frame from those of facial features. This approach does not require any prior knowledge about face pose, eye positions, or the shape of the glasses.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115314804","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048489
E. Tassone, G. West, S. Venkatesh
Gait classification is a developing research area, particularly with regards to biometrics. It aims to use the distinctive spatial and temporal characteristics of human motion to classify differing activities. As a biometric, this extends to recognising different people by the heterogeneous aspects of their gait. This research aims to use a modified deformable model, the temporal PDM, to distinguish the movements of a walking and running person. The movement of 2D points on the moving form is used to provide input into the model and classify the type of gait present.
{"title":"Temporal PDMs for gait classification","authors":"E. Tassone, G. West, S. Venkatesh","doi":"10.1109/ICPR.2002.1048489","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048489","url":null,"abstract":"Gait classification is a developing research area, particularly with regards to biometrics. It aims to use the distinctive spatial and temporal characteristics of human motion to classify differing activities. As a biometric, this extends to recognising different people by the heterogeneous aspects of their gait. This research aims to use a modified deformable model, the temporal PDM, to distinguish the movements of a walking and running person. The movement of 2D points on the moving form is used to provide input into the model and classify the type of gait present.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116657848","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048406
Zhen Ye, Cheng-Chang Lu
A new texture image segmentation algorithm, HMTseg, was recently proposed and applied successfully to supervised segmentation. In this paper, we extend the HMTseg algorithm to unsupervised SAR image segmentation. A multiscale Expectation Maximization (EM) algorithm is used to integrate the parameter estimation and classification into one. Because of the high levels of speckle noise present at fine scales in SAR images, segmentations on coarse scales are more reliable and accurate than those on fine scales. Based on the Hybrid Contextual Labelling Tree (HCLT) model, a weight factor /spl beta/, is introduced to increase the emphasis of context information. Ultimately, a Bayesian interscale and intrascale fusion algorithm is applied to refine raw segmentations.
{"title":"Wavelet-based unsupervised SAR image segmentation using hidden Markov tree models","authors":"Zhen Ye, Cheng-Chang Lu","doi":"10.1109/ICPR.2002.1048406","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048406","url":null,"abstract":"A new texture image segmentation algorithm, HMTseg, was recently proposed and applied successfully to supervised segmentation. In this paper, we extend the HMTseg algorithm to unsupervised SAR image segmentation. A multiscale Expectation Maximization (EM) algorithm is used to integrate the parameter estimation and classification into one. Because of the high levels of speckle noise present at fine scales in SAR images, segmentations on coarse scales are more reliable and accurate than those on fine scales. Based on the Hybrid Contextual Labelling Tree (HCLT) model, a weight factor /spl beta/, is introduced to increase the emphasis of context information. Ultimately, a Bayesian interscale and intrascale fusion algorithm is applied to refine raw segmentations.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123812499","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048469
H. Wildenauer, T. Melzer, H. Bischof
In the recent literature, gradient-based (filtered) eigenspaces have been used as a means to achieve illumination insensitivity. In this paper we show that filtered eigenspaces are also inherently robust w.r.t. (non-Gaussian) noise and occlusions. We argue that this robustness stems essentially from the sparseness of representation and insensitivity w.r.t. shifts in the mean value. This is also demonstrated experimentally using examples from the field of object recognition and pose estimation.
{"title":"A gradient-based eigenspace approach to dealing with occlusions and non-Gaussian noise","authors":"H. Wildenauer, T. Melzer, H. Bischof","doi":"10.1109/ICPR.2002.1048469","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048469","url":null,"abstract":"In the recent literature, gradient-based (filtered) eigenspaces have been used as a means to achieve illumination insensitivity. In this paper we show that filtered eigenspaces are also inherently robust w.r.t. (non-Gaussian) noise and occlusions. We argue that this robustness stems essentially from the sparseness of representation and insensitivity w.r.t. shifts in the mean value. This is also demonstrated experimentally using examples from the field of object recognition and pose estimation.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122782054","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048248
K. Torkkola
We argue that optimal feature selection is intrinsically a harder problem than learning discriminative feature transforms, provided a suitable criterion for the latter. We discuss mutual information between class labels and transformed features as such a criterion. Instead of Shannon's definition we use measures based on Renyi entropy, which lends itself into an efficient implementation and an interpretation of "information forces" induced by samples of data that drive the transform.
{"title":"Learning feature transforms is an easier problem than feature selection","authors":"K. Torkkola","doi":"10.1109/ICPR.2002.1048248","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048248","url":null,"abstract":"We argue that optimal feature selection is intrinsically a harder problem than learning discriminative feature transforms, provided a suitable criterion for the latter. We discuss mutual information between class labels and transformed features as such a criterion. Instead of Shannon's definition we use measures based on Renyi entropy, which lends itself into an efficient implementation and an interpretation of \"information forces\" induced by samples of data that drive the transform.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123031852","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 : 2002-12-10DOI: 10.1109/ICPR.2002.1048441
H. Cardot, O. Lézoray
This paper presents a new architecture of neural networks designed for pattern recognition. The concept of induction graphs coupled with a divide-and-conquer strategy defines a Graph of Neural Network (GNN). It is based on a set of several little neural networks, each one discriminating only two classes. The principles used to perform the decision of classification are : a branch quality index and a selection by elimination. A significant gain in the global classification rate can be obtained by using a GNN. This is illustrated by tests on databases from the UCI machine learning database repository. The experimental results show that a GNN can achieve an improved performance in classification.
{"title":"Graph of neural networks for pattern recognition","authors":"H. Cardot, O. Lézoray","doi":"10.1109/ICPR.2002.1048441","DOIUrl":"https://doi.org/10.1109/ICPR.2002.1048441","url":null,"abstract":"This paper presents a new architecture of neural networks designed for pattern recognition. The concept of induction graphs coupled with a divide-and-conquer strategy defines a Graph of Neural Network (GNN). It is based on a set of several little neural networks, each one discriminating only two classes. The principles used to perform the decision of classification are : a branch quality index and a selection by elimination. A significant gain in the global classification rate can be obtained by using a GNN. This is illustrated by tests on databases from the UCI machine learning database repository. The experimental results show that a GNN can achieve an improved performance in classification.","PeriodicalId":159502,"journal":{"name":"Object recognition supported by user interaction for service robots","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131263343","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：