This paper presents a new and fast algorithm for multi-baseline stereo designed to handle the occlusion problem. The algorithm is a hybrid between fast heuristic occlusion overcoming algorithms that precompute an approximate visibility and slower methods that use correct visibility handling. Our approach is based on iterative dynamic programming and computes simultaneously disparity and camera visibility. Interestingly, dynamic programming makes it possible to compute exactly part of the visibility information. The remainder is obtained through heuristics. The validity of our scheme is established using real imagery with ground truth and compares favorably with other state-of-the-art multi-baseline stereo algorithms.
{"title":"Fast multiple-baseline stereo with occlusion","authors":"M. Drouin, Martin Trudeau, S. Roy","doi":"10.1109/3DIM.2005.40","DOIUrl":"https://doi.org/10.1109/3DIM.2005.40","url":null,"abstract":"This paper presents a new and fast algorithm for multi-baseline stereo designed to handle the occlusion problem. The algorithm is a hybrid between fast heuristic occlusion overcoming algorithms that precompute an approximate visibility and slower methods that use correct visibility handling. Our approach is based on iterative dynamic programming and computes simultaneously disparity and camera visibility. Interestingly, dynamic programming makes it possible to compute exactly part of the visibility information. The remainder is obtained through heuristics. The validity of our scheme is established using real imagery with ground truth and compares favorably with other state-of-the-art multi-baseline stereo algorithms.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125977200","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}
Toward a successful 3D and textural reconstruction of urban scenes, the use of both single-row based telemetric and photographic data in a same framework has proved to be a powerful technique. A necessary condition to obtain good results is to accurately calibrate the telemetric and photographic sensors together. We present a study of this calibration process and propose an improved extrinsic calibration technique. It is based on an existing technique which consists in scanning a planar pattern in several poses, giving a set of relative position and orientation constraints. The innovation is the use of a more appropriate laser beam distance between telemetric points and the planar target. Moreover, we use robust methods to manage outliers at several steps of the algorithm. Improved results on both theoretical and experimental data are given.
{"title":"An improved calibration technique for coupled single-row telemeter and CCD camera","authors":"R. Dupont, R. Keriven, P. Fuchs","doi":"10.1109/3DIM.2005.19","DOIUrl":"https://doi.org/10.1109/3DIM.2005.19","url":null,"abstract":"Toward a successful 3D and textural reconstruction of urban scenes, the use of both single-row based telemetric and photographic data in a same framework has proved to be a powerful technique. A necessary condition to obtain good results is to accurately calibrate the telemetric and photographic sensors together. We present a study of this calibration process and propose an improved extrinsic calibration technique. It is based on an existing technique which consists in scanning a planar pattern in several poses, giving a set of relative position and orientation constraints. The innovation is the use of a more appropriate laser beam distance between telemetric points and the planar target. Moreover, we use robust methods to manage outliers at several steps of the algorithm. Improved results on both theoretical and experimental data are given.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128325960","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 aperture problem is one of the omnipresent issues in computer vision. Its local character constrains point matching to high textured areas, so that points in gradient-oriented regions (such as straight lines) can not be reliably matched. We propose a new method to overcome this problem by devising a global matching strategy under the factorization framework. We solve the n-frame correspondence problem under this context by assuming the rigidity of the scene. To this end, a geometric constraint is used that selects the matching solution resulting in a rank-4 observation matrix. The rank of the observation matrix is a function of the matching solutions associated to each image and as such a simultaneous solution for all frames has to be found. An optimization procedure is used in this text in order to find the solution.
{"title":"Contour point tracking by enforcement of rigidity constraints","authors":"Ricardo Oliveira, J. Costeira, J. Xavier","doi":"10.1109/3DIM.2005.27","DOIUrl":"https://doi.org/10.1109/3DIM.2005.27","url":null,"abstract":"The aperture problem is one of the omnipresent issues in computer vision. Its local character constrains point matching to high textured areas, so that points in gradient-oriented regions (such as straight lines) can not be reliably matched. We propose a new method to overcome this problem by devising a global matching strategy under the factorization framework. We solve the n-frame correspondence problem under this context by assuming the rigidity of the scene. To this end, a geometric constraint is used that selects the matching solution resulting in a rank-4 observation matrix. The rank of the observation matrix is a function of the matching solutions associated to each image and as such a simultaneous solution for all frames has to be found. An optimization procedure is used in this text in order to find the solution.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131607048","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 present a flexible modeling system for obtaining the texture-mapped 3D geometric model. The modeling system uses an algorithm combining shape-from-silhouette with light-sectioning. In the algorithm, at first, a rough shape model is obtained by shape-from-silhouette method almost automatically. Next, concavities and complex parts on the object surface are obtained by light-sectioning method with manual scanning. For applying light-sectioning method to volume data, we propose volumetric light-sectioning algorithm. Then our modeling system can realize easy and accurate generation of 3D geometric model.
{"title":"A flexible 3D modeling system based on combining shape-from-silhouette with light-sectioning algorithm","authors":"T. Terauchi, Y. Oue, K. Fujimura","doi":"10.1109/3DIM.2005.8","DOIUrl":"https://doi.org/10.1109/3DIM.2005.8","url":null,"abstract":"In this paper we present a flexible modeling system for obtaining the texture-mapped 3D geometric model. The modeling system uses an algorithm combining shape-from-silhouette with light-sectioning. In the algorithm, at first, a rough shape model is obtained by shape-from-silhouette method almost automatically. Next, concavities and complex parts on the object surface are obtained by light-sectioning method with manual scanning. For applying light-sectioning method to volume data, we propose volumetric light-sectioning algorithm. Then our modeling system can realize easy and accurate generation of 3D geometric model.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126233185","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}
We investigate the possibility of Euclidean reconstruction from translational motion, using multiple uncalibrated cameras. We show that in the case of multiple cameras viewing a translating scene, no additional constraints are given by the translational motion compared to the more general case with one camera viewing a scene undergoing a general motion. However, the knowledge of translational motion allows an intermediate affine reconstruction from each camera, and aids in the reconstruction process by simplifying several steps, resulting in a more reliable algorithm for 3D reconstruction. We also identify the critical directions of translation, for which no affine reconstruction is possible. Experiments on real and simulated data are performed to illustrate that the method works in practice.
{"title":"Euclidean reconstruction from translational motion using multiple cameras","authors":"Pär Hammarstedt, A. Heyden","doi":"10.1109/3DIM.2005.36","DOIUrl":"https://doi.org/10.1109/3DIM.2005.36","url":null,"abstract":"We investigate the possibility of Euclidean reconstruction from translational motion, using multiple uncalibrated cameras. We show that in the case of multiple cameras viewing a translating scene, no additional constraints are given by the translational motion compared to the more general case with one camera viewing a scene undergoing a general motion. However, the knowledge of translational motion allows an intermediate affine reconstruction from each camera, and aids in the reconstruction process by simplifying several steps, resulting in a more reliable algorithm for 3D reconstruction. We also identify the critical directions of translation, for which no affine reconstruction is possible. Experiments on real and simulated data are performed to illustrate that the method works in practice.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125691661","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}
A mechanism is introduced that automatically integrates multi-view range images without registering the images. The mechanism is based on a reference double-frame that acts as the coordinate system of the scene. A single-view range image of a scene is obtained by sweeping a laser line over the scene by hand and analyzing the acquired light stripes. Range images captured from different views of the scene are in the coordinate system of the double-frame, and thus, automatically integrate without further processing.
{"title":"A mechanism for range image integration without image registration","authors":"L. Zagorchev, A. Goshtasby","doi":"10.1109/3DIM.2005.10","DOIUrl":"https://doi.org/10.1109/3DIM.2005.10","url":null,"abstract":"A mechanism is introduced that automatically integrates multi-view range images without registering the images. The mechanism is based on a reference double-frame that acts as the coordinate system of the scene. A single-view range image of a scene is obtained by sweeping a laser line over the scene by hand and analyzing the acquired light stripes. Range images captured from different views of the scene are in the coordinate system of the double-frame, and thus, automatically integrate without further processing.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133859540","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 present an approach that is able to reconstruct 3D models from extended video sequences captured with an uncalibrated hand-held camera. We focus on two specific issues: (1) key-frame selection; and (2) projective drift. Given a long video sequence it is often not practical to work with all video frames. In addition, to allow for effective outlier rejection and motion estimation it is necessary to have a sufficient baseline between frames. For this purpose, we propose a key-frame selection procedure based on a robust model selection criterion. Our approach guarantees that the camera motion can be estimated reliably by analyzing the feature correspondences between three consecutive views. Another problem for long uncalibrated video sequences is projective drift. Error accumulation leads to a non-projective distortion of the model. This causes the projective basis at the beginning and the end of the sequence to become inconsistent and leads to the failure of self-calibration. We propose a self-calibration approach that is insensitive to this global projective drift. After self-calibration triplets of key-frames are aligned using absolute orientation and hierarchically merged into a complete metric reconstruction. Next, we compute a detailed 3D surface model using stereo matching. The 3D model is textured using some of the frames.
{"title":"3D models from extended uncalibrated video sequences: addressing key-frame selection and projective drift","authors":"Jason Repko, M. Pollefeys","doi":"10.1109/3dim.2005.4","DOIUrl":"https://doi.org/10.1109/3dim.2005.4","url":null,"abstract":"In this paper, we present an approach that is able to reconstruct 3D models from extended video sequences captured with an uncalibrated hand-held camera. We focus on two specific issues: (1) key-frame selection; and (2) projective drift. Given a long video sequence it is often not practical to work with all video frames. In addition, to allow for effective outlier rejection and motion estimation it is necessary to have a sufficient baseline between frames. For this purpose, we propose a key-frame selection procedure based on a robust model selection criterion. Our approach guarantees that the camera motion can be estimated reliably by analyzing the feature correspondences between three consecutive views. Another problem for long uncalibrated video sequences is projective drift. Error accumulation leads to a non-projective distortion of the model. This causes the projective basis at the beginning and the end of the sequence to become inconsistent and leads to the failure of self-calibration. We propose a self-calibration approach that is insensitive to this global projective drift. After self-calibration triplets of key-frames are aligned using absolute orientation and hierarchically merged into a complete metric reconstruction. Next, we compute a detailed 3D surface model using stereo matching. The 3D model is textured using some of the frames.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"340 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124780411","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}
Existing techniques for view-based 3D object recognition using computer vision rely on training the system on a particular object before it is introduced into an environment. This training often consists of taking over 100 images at predetermined points around the viewing sphere in an attempt to account for most angles for viewing the object. However, in many circumstances, the environment is well known and we only expect to see a small subset of all possible appearances. In this paper, we test the idea that under these conditions, it is possible to train an object recognition system on-the-fly using images of an object as it appears in its environment, with supervision from the user. Furthermore, because some views of an object are much more likely than others, the number of training images required can be greatly reduced.
{"title":"Acquisition of view-based 3D object models using supervised, unstructured data","authors":"Kevin Coogan, I. Green","doi":"10.1109/3DIM.2005.15","DOIUrl":"https://doi.org/10.1109/3DIM.2005.15","url":null,"abstract":"Existing techniques for view-based 3D object recognition using computer vision rely on training the system on a particular object before it is introduced into an environment. This training often consists of taking over 100 images at predetermined points around the viewing sphere in an attempt to account for most angles for viewing the object. However, in many circumstances, the environment is well known and we only expect to see a small subset of all possible appearances. In this paper, we test the idea that under these conditions, it is possible to train an object recognition system on-the-fly using images of an object as it appears in its environment, with supervision from the user. Furthermore, because some views of an object are much more likely than others, the number of training images required can be greatly reduced.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128631600","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 review several different 3D scanning devices. We present a method for empirical accuracy analysis, and apply it to several scanners providing an overview of their technologies. The scanners include both general purpose and face specific scanning devices. We focus on face scanning technique, although the technique should be applicable to other domains as well. The proposed method involves several different calibration faces of known shape and comparisons of their scans to investigate both absolute accuracy and repeatability.
{"title":"Accuracy of 3D scanning technologies in a face scanning scenario","authors":"Chris Boehnen, P. Flynn","doi":"10.1109/3DIM.2005.13","DOIUrl":"https://doi.org/10.1109/3DIM.2005.13","url":null,"abstract":"In this paper, we review several different 3D scanning devices. We present a method for empirical accuracy analysis, and apply it to several scanners providing an overview of their technologies. The scanners include both general purpose and face specific scanning devices. We focus on face scanning technique, although the technique should be applicable to other domains as well. The proposed method involves several different calibration faces of known shape and comparisons of their scans to investigate both absolute accuracy and repeatability.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121949729","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 presents a novel approach for view synthesis and image interpolation. The algorithm is build up in a hierarchical way, and this on different structural levels instead of using a classic image pyramid. First coarse matching is done on a 'shape basis' only. A background-foreground segmentation yields a fairly accurate contour for every incoming video stream. Inter-relating these contours is a 1D problem and as such very fast. This step is then used to compute small position dependent bounding-boxes in 3D space which enclose the underlying object. The next step is a more expensive window based matching, within the volume of these bounding-boxes. This is limited to a number of regions around 'promising' feature points. Global regularisation is obtained by a graph cut. Speed results here from limiting the number of feature points. In a third step the interpolation is 'pre-rendered' and simultaneously evaluated on a per pixel basis. This is done by computing a Birchfield dissimilarity measure on the GPU. Per pixel parallelised operations keep computational cost low. Finally the bad interpolated parts are 'patched'. This per pixel correction yields the final interpolated view at the finest level. Here we also deal explicitly with opacity at the borders of the foreground object.
{"title":"Hierarchical coarse to fine depth estimation for realistic view interpolation","authors":"I. Geys, L. Gool","doi":"10.1109/3DIM.2005.52","DOIUrl":"https://doi.org/10.1109/3DIM.2005.52","url":null,"abstract":"This paper presents a novel approach for view synthesis and image interpolation. The algorithm is build up in a hierarchical way, and this on different structural levels instead of using a classic image pyramid. First coarse matching is done on a 'shape basis' only. A background-foreground segmentation yields a fairly accurate contour for every incoming video stream. Inter-relating these contours is a 1D problem and as such very fast. This step is then used to compute small position dependent bounding-boxes in 3D space which enclose the underlying object. The next step is a more expensive window based matching, within the volume of these bounding-boxes. This is limited to a number of regions around 'promising' feature points. Global regularisation is obtained by a graph cut. Speed results here from limiting the number of feature points. In a third step the interpolation is 'pre-rendered' and simultaneously evaluated on a per pixel basis. This is done by computing a Birchfield dissimilarity measure on the GPU. Per pixel parallelised operations keep computational cost low. Finally the bad interpolated parts are 'patched'. This per pixel correction yields the final interpolated view at the finest level. Here we also deal explicitly with opacity at the borders of the foreground object.","PeriodicalId":170883,"journal":{"name":"Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127056179","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：