We propose a novel procedure for estimating blur in a single image corrupted by blur due to a rigid camera motion during the exposure. Often this blur is approximated as space invariant, even if this assumption holds, for example, only on small image region in perspective images captured during camera movement. Our algorithm analyzes separately selected image regions containing a corner and in each region the blur is described by its direction and extent. The algorithm works directly in space domain, exploiting gradient vectors at pixels belonging to the blurred corner edges. The algorithm has been successfully tested both on synthetic and real images showing good performance even on small image regions and in presence of noise.
{"title":"Corner Displacement from Motion Blur","authors":"G. Boracchi, V. Caglioti","doi":"10.1109/ICIAP.2007.47","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.47","url":null,"abstract":"We propose a novel procedure for estimating blur in a single image corrupted by blur due to a rigid camera motion during the exposure. Often this blur is approximated as space invariant, even if this assumption holds, for example, only on small image region in perspective images captured during camera movement. Our algorithm analyzes separately selected image regions containing a corner and in each region the blur is described by its direction and extent. The algorithm works directly in space domain, exploiting gradient vectors at pixels belonging to the blurred corner edges. The algorithm has been successfully tested both on synthetic and real images showing good performance even on small image regions and in presence of noise.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"49 21","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113936557","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}
R. Ewerth, M. Schwalb, P. Tessmann, Bernd Freisleben
The distinction of translational and rotational camera motion and the recognition of moving objects is an important topic for scientific film studies. In this paper, we present an approach to distinguish between camera and object motion in MPEG videos and provide a pixel-accurate segmentation of moving objects. Compressed domain features are used as far as possible in order to reduce computation time. First, camera motion parameters are estimated and translational movements are distinguished from rotational movements based on a three-dimensional (3D) camera model. Then, motion vectors which do not fit to the camera motion estimate are assigned to object clusters. The moving object information is utilized to refine the camera motion estimate, and a novel compressed domain tracking algorithm is applied to verify the temporal consistency of detected objects. In contrast to previous approaches, the tracking of both moving objects and background allows to perform their separation iteratively only once per shot. The object boundary is estimated with pixel accuracy via active contour models. Experimental results demonstrate the feasibility of the proposed algorithm.
{"title":"Segmenting Moving Objects in MPEG Videos in the Presence of Camera Motion","authors":"R. Ewerth, M. Schwalb, P. Tessmann, Bernd Freisleben","doi":"10.1109/ICIAP.2007.115","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.115","url":null,"abstract":"The distinction of translational and rotational camera motion and the recognition of moving objects is an important topic for scientific film studies. In this paper, we present an approach to distinguish between camera and object motion in MPEG videos and provide a pixel-accurate segmentation of moving objects. Compressed domain features are used as far as possible in order to reduce computation time. First, camera motion parameters are estimated and translational movements are distinguished from rotational movements based on a three-dimensional (3D) camera model. Then, motion vectors which do not fit to the camera motion estimate are assigned to object clusters. The moving object information is utilized to refine the camera motion estimate, and a novel compressed domain tracking algorithm is applied to verify the temporal consistency of detected objects. In contrast to previous approaches, the tracking of both moving objects and background allows to perform their separation iteratively only once per shot. The object boundary is estimated with pixel accuracy via active contour models. Experimental results demonstrate the feasibility of the proposed algorithm.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133879321","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 evaluate the performance of text- independent writer identification methods on a handwriting dataset containing medieval English documents. Applicable identification rates are achieved by combining textural features (joint directional probability distributions) with allographic features (grapheme-emission distributions). The aim is to develop an automatic handwriting identification tool that can assist the paleographer in the task of determining the authorship of historical manuscripts.
{"title":"Automatic Handwriting Identification on Medieval Documents","authors":"M. Bulacu, Lambert Schomaker","doi":"10.1109/ICIAP.2007.33","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.33","url":null,"abstract":"In this paper, we evaluate the performance of text- independent writer identification methods on a handwriting dataset containing medieval English documents. Applicable identification rates are achieved by combining textural features (joint directional probability distributions) with allographic features (grapheme-emission distributions). The aim is to develop an automatic handwriting identification tool that can assist the paleographer in the task of determining the authorship of historical manuscripts.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"336 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115456470","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}
Chauã C. Queirolo, Maurício Pamplona Segundo, O. Bellon, Luciano Silva
This paper presents a new method for 3D face recognition. The method combines a Simulated Annealing-based approach for image registration using the surface interpenetration measure (SIM) to perform a precise matching between two face images. The recognition score is obtained by combining the SIM scores of four different face regions after their alignment. Experiments were conducted on two databases with a variety official expressions. The images from the databases were classified according to noise level and facial expression, allowing the analysis of each particular effect on 3D face recognition. The method allows a verification rate of 99.9%, at a false acceptance rate (FAR) of 0%, for the FRGC ver 2.0 database when only noiseless, neutral expression face images are used. Also, the results using face images with expressions and noise demonstrate that subjects still can be recognized with 87.5% of verification rate, at a FAR of 0%.
{"title":"Noise versus Facial Expression on 3D Face Recognition","authors":"Chauã C. Queirolo, Maurício Pamplona Segundo, O. Bellon, Luciano Silva","doi":"10.1109/ICIAP.2007.94","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.94","url":null,"abstract":"This paper presents a new method for 3D face recognition. The method combines a Simulated Annealing-based approach for image registration using the surface interpenetration measure (SIM) to perform a precise matching between two face images. The recognition score is obtained by combining the SIM scores of four different face regions after their alignment. Experiments were conducted on two databases with a variety official expressions. The images from the databases were classified according to noise level and facial expression, allowing the analysis of each particular effect on 3D face recognition. The method allows a verification rate of 99.9%, at a false acceptance rate (FAR) of 0%, for the FRGC ver 2.0 database when only noiseless, neutral expression face images are used. Also, the results using face images with expressions and noise demonstrate that subjects still can be recognized with 87.5% of verification rate, at a FAR of 0%.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115571553","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 Scale Invariant Feature Transform, SIFT, has been successfully applied to robot vision, object recognition, motion estimation, etc. In this work, we propose a SIFT improvement that makes feature extraction and matching more robust, adding a feature group matching layer, which takes into account mutual spatial relations between features. The feature group matching is very fast to be computed and leads to interesting results, above all for the absence of outliers. Results of vision based robot localization using the proposed approach are presented.
{"title":"Visual feature group matching for autonomous robot localization","authors":"E. Frontoni, P. Zingaretti","doi":"10.1109/ICIAP.2007.137","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.137","url":null,"abstract":"The Scale Invariant Feature Transform, SIFT, has been successfully applied to robot vision, object recognition, motion estimation, etc. In this work, we propose a SIFT improvement that makes feature extraction and matching more robust, adding a feature group matching layer, which takes into account mutual spatial relations between features. The feature group matching is very fast to be computed and leads to interesting results, above all for the absence of outliers. Results of vision based robot localization using the proposed approach are presented.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123895172","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 introduces a framework for the generation of genuine connectivity relations whose equivalent classes (called connected components) define unique partitions of the definition domain of a given grey tone image. This framework exploits the total ordering relation between the alpha-connected components of a pixel (two pixels are alpha-connected if there exists at least one path joining them such that the intensity differences between successive pixels of this path does not exceed a threshold value equal to alpha). Genuine connectivity relations are then obtained by considering the largest alpha-connected components satisfying one or more logical predicates such as the variance of the intensity values of the alpha-connected components not exceeding a given threshold value. Fine to coarse hierarchy partitions are generated by carefully varying the input threshold values. The proposed framework has the striking property of uniqueness. That is, the results do not depend on pixel processing order and are fully defined by the values of the threshold values, in contrast to most region growing procedures.
{"title":"On Genuine Connectivity Relations Based on Logical Predicates","authors":"P. Soille","doi":"10.1109/ICIAP.2007.96","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.96","url":null,"abstract":"This paper introduces a framework for the generation of genuine connectivity relations whose equivalent classes (called connected components) define unique partitions of the definition domain of a given grey tone image. This framework exploits the total ordering relation between the alpha-connected components of a pixel (two pixels are alpha-connected if there exists at least one path joining them such that the intensity differences between successive pixels of this path does not exceed a threshold value equal to alpha). Genuine connectivity relations are then obtained by considering the largest alpha-connected components satisfying one or more logical predicates such as the variance of the intensity values of the alpha-connected components not exceeding a given threshold value. Fine to coarse hierarchy partitions are generated by carefully varying the input threshold values. The proposed framework has the striking property of uniqueness. That is, the results do not depend on pixel processing order and are fully defined by the values of the threshold values, in contrast to most region growing procedures.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115051277","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 tensor voting framework (TVF), proposed by Medioni at at, has proved its effectiveness in perceptual grouping of arbitrary dimensional data. In the computer vision and image processing fields, this algorithm has been applied to solve various problems like stereo-matching, 3D reconstruction, and image in painting. The TVF technique can detect and remove a big percentage of outliers, but unfortunately it does not generate satisfactory results when the data are corrupted by additive noise. In this paper a new direct votes computation algorithm for high dimensional spaces is described, and a parametric class of decay functions is proposed to deal with noisy data. Preliminary comparative results between the original TVF and our algorithm are shown on synthetic data.
{"title":"Tensor Voting Fields: Direct Votes Computation and New Saliency Functions","authors":"P. Campadelli, G. Lombardi","doi":"10.1109/ICIAP.2007.124","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.124","url":null,"abstract":"The tensor voting framework (TVF), proposed by Medioni at at, has proved its effectiveness in perceptual grouping of arbitrary dimensional data. In the computer vision and image processing fields, this algorithm has been applied to solve various problems like stereo-matching, 3D reconstruction, and image in painting. The TVF technique can detect and remove a big percentage of outliers, but unfortunately it does not generate satisfactory results when the data are corrupted by additive noise. In this paper a new direct votes computation algorithm for high dimensional spaces is described, and a parametric class of decay functions is proposed to deal with noisy data. Preliminary comparative results between the original TVF and our algorithm are shown on synthetic data.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128374318","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 projector-camera system has recently been studied extensively as one of new information presenting systems. For generating screen images properly, it is important to calibrate projector-camera systems accurately. The existing methods for calibrating projector-camera systems are based on 4 markers on the screen and 4 light projections from projectors, and thus require at least 8 basis points totally in images. However, it is not easy to track 8 or more basis points reliably in images, if the projector camera system moves arbitrarily. Thus, we in this paper propose a method for generating screen images properly from less basis points in camera images.
{"title":"Calibration and Image Generation of Mobile Projector-Camera Systems","authors":"K. Hamada, J. Sato","doi":"10.1109/ICIAP.2007.39","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.39","url":null,"abstract":"The projector-camera system has recently been studied extensively as one of new information presenting systems. For generating screen images properly, it is important to calibrate projector-camera systems accurately. The existing methods for calibrating projector-camera systems are based on 4 markers on the screen and 4 light projections from projectors, and thus require at least 8 basis points totally in images. However, it is not easy to track 8 or more basis points reliably in images, if the projector camera system moves arbitrarily. Thus, we in this paper propose a method for generating screen images properly from less basis points in camera images.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129803128","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 deals with the process of view synthesis based on the relative affine structure. It describes a complete pipeline that, starting with uncalibrated images, produces a virtual sequence with viewpoint control. Experiments illustrate the approach.
{"title":"An Uncalibrated View-Synthesis Pipeline","authors":"Andrea Fusiello, L. Irsara","doi":"10.1109/ICIAP.2007.24","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.24","url":null,"abstract":"This paper deals with the process of view synthesis based on the relative affine structure. It describes a complete pipeline that, starting with uncalibrated images, produces a virtual sequence with viewpoint control. Experiments illustrate the approach.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125652923","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 : 2007-09-10DOI: 10.1093/ietisy/e91-d.8.2171
Ying Piao, J. Sato
Recently, many application systems have been developed by using a large number of cameras. If the 3D points are observed from synchronized cameras, the multiple view geometry of these cameras can be computed and the 3D reconstruction of the scene is available. Thus, the synchronization of multiple cameras is essential. In this paper, we propose a method for finding synchronization of multiple cameras and for computing the epipolar geometry from un- calibrated and unsynchronized cameras. In particular we use the affine invariance on frame numbers of camera images for finding the synchronization. The proposed method is tested by using real image sequences taken from uncalibrated and unsynchronized cameras.
{"title":"Computing Epipolar Geometry from Unsynchronized Cameras","authors":"Ying Piao, J. Sato","doi":"10.1093/ietisy/e91-d.8.2171","DOIUrl":"https://doi.org/10.1093/ietisy/e91-d.8.2171","url":null,"abstract":"Recently, many application systems have been developed by using a large number of cameras. If the 3D points are observed from synchronized cameras, the multiple view geometry of these cameras can be computed and the 3D reconstruction of the scene is available. Thus, the synchronization of multiple cameras is essential. In this paper, we propose a method for finding synchronization of multiple cameras and for computing the epipolar geometry from un- calibrated and unsynchronized cameras. In particular we use the affine invariance on frame numbers of camera images for finding the synchronization. The proposed method is tested by using real image sequences taken from uncalibrated and unsynchronized cameras.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130674001","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}