Pub Date : 1994-11-11DOI: 10.1109/MNRAO.1994.346250
N. H. Goddard
Discrimination of articulated movement is a central problem in perception. We present a model-based system designed to discriminate articulated movements and demonstrate its capabilities in distinguishing three human gaits. Recognition proceeds directly from uninterpreted visual motion features. The adaptive model of movement, the scenario, is based on discrete parameterized events separated by parameterized time intervals. Experimental results are shown for real data derived from moving light displays, demonstrating the effectiveness of the structured connectionist approach.<>
{"title":"Incremental model-based discrimination of articulated movement from motion features","authors":"N. H. Goddard","doi":"10.1109/MNRAO.1994.346250","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346250","url":null,"abstract":"Discrimination of articulated movement is a central problem in perception. We present a model-based system designed to discriminate articulated movements and demonstrate its capabilities in distinguishing three human gaits. Recognition proceeds directly from uninterpreted visual motion features. The adaptive model of movement, the scenario, is based on discrete parameterized events separated by parameterized time intervals. Experimental results are shown for real data derived from moving light displays, demonstrating the effectiveness of the structured connectionist approach.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"147 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122915293","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346243
C. Nastar, N. Ayache
Presents a method for analysis of nonrigid motion in time sequences of volume images (4D data). In this method, nonrigid motion of the deforming object's contour is dynamically approximated by a deformable surface. In order to reduce the number of parameters describing the deformation, we make use of modal analysis, which provides a spatial smoothing of the surface, and Fourier analysis of the time signals of the main deformation spectrum components, which provides a temporal smoothing. Thus, a complex nonrigid deformation displayed in 4D data is described by very few parameters: the main excited spatial modes and the main Fourier harmonics. Therefore, 4D data can be analyzed and reduced in a very efficient way. The power of the approach is illustrated by results on 4D heart-scan data.<>
{"title":"Spatio-temporal analysis of nonrigid motion from 4D data","authors":"C. Nastar, N. Ayache","doi":"10.1109/MNRAO.1994.346243","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346243","url":null,"abstract":"Presents a method for analysis of nonrigid motion in time sequences of volume images (4D data). In this method, nonrigid motion of the deforming object's contour is dynamically approximated by a deformable surface. In order to reduce the number of parameters describing the deformation, we make use of modal analysis, which provides a spatial smoothing of the surface, and Fourier analysis of the time signals of the main deformation spectrum components, which provides a temporal smoothing. Thus, a complex nonrigid deformation displayed in 4D data is described by very few parameters: the main excited spatial modes and the main Fourier harmonics. Therefore, 4D data can be analyzed and reduced in a very efficient way. The power of the approach is illustrated by results on 4D heart-scan data.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133818398","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346233
C. Gear
We address the problem of grouping points or features common to a single object. In this paper we consider the processing of a sequence of two-dimensional orthogonal projections of a three-dimensional scene containing an unknown number of independently-moving rigid objects. We describe a computationally inexpensive algorithm that can determine the number of bodies and which points belong to which body.<>
{"title":"Feature grouping in moving objects","authors":"C. Gear","doi":"10.1109/MNRAO.1994.346233","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346233","url":null,"abstract":"We address the problem of grouping points or features common to a single object. In this paper we consider the processing of a sequence of two-dimensional orthogonal projections of a three-dimensional scene containing an unknown number of independently-moving rigid objects. We describe a computationally inexpensive algorithm that can determine the number of bodies and which points belong to which body.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117039030","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346248
H. Sawhney
Presents a formulation for 3D motion and structure analysis using motion parallax defined with respect to an arbitrary plane in the environment. It is shown that if an image coordinate system is warped using plane projective transformation with respect to a reference view, the residual image motion is dependent only on the epipoles and has a simple relation to the 3D structure. Our computational scheme avoids point/line correspondence and is based on hierarchical estimation and image warping working directly with spatiotemporal image intensities. Results on real images demonstrate how this analysis can be used to simplify ego and object motion segmentation.<>
{"title":"Simplifying multiple motion and structure analysis using planar parallax and image warping","authors":"H. Sawhney","doi":"10.1109/MNRAO.1994.346248","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346248","url":null,"abstract":"Presents a formulation for 3D motion and structure analysis using motion parallax defined with respect to an arbitrary plane in the environment. It is shown that if an image coordinate system is warped using plane projective transformation with respect to a reference view, the residual image motion is dependent only on the epipoles and has a simple relation to the 3D structure. Our computational scheme avoids point/line correspondence and is based on hierarchical estimation and image warping working directly with spatiotemporal image intensities. Results on real images demonstrate how this analysis can be used to simplify ego and object motion segmentation.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127097166","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346253
Sourabh A. Niyogi, E. Adelson
Human motions generate characteristic spatiotemporal patterns. We have developed a set of techniques for analyzing the patterns generated by people walking across the field of view. After change detection, the XYT pattern can be fit with a smooth spatiotemporal surface. This surface is approximately periodic, reflecting the periodicity of the gait. The surface can be expressed as a combination of a standard parameterized surface-the canonical walk-and a deviation surface that is specific to the individual walk.<>
{"title":"Analyzing gait with spatiotemporal surfaces","authors":"Sourabh A. Niyogi, E. Adelson","doi":"10.1109/MNRAO.1994.346253","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346253","url":null,"abstract":"Human motions generate characteristic spatiotemporal patterns. We have developed a set of techniques for analyzing the patterns generated by people walking across the field of view. After change detection, the XYT pattern can be fit with a smooth spatiotemporal surface. This surface is approximately periodic, reflecting the periodicity of the gait. The surface can be expressed as a combination of a standard parameterized surface-the canonical walk-and a deviation surface that is specific to the individual walk.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116860901","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346252
A. G. Bharatkumar, K. E. Daigle, Marcus G. Pandy, Q. Cai, Jake K. Aggarwal
The paper describes a simple model for free-speed human walking and compares ordinary images of a walking person to the model. Three dimensional kinematic data were obtained from subjects walking with markers over the joints of their limbs. The average of these data was used to derive a model stick figure of the lower limbs, based on the average anthropometric data of the population. Stick figures were obtained from ordinary images of persons dressed in tight fitting clothes without any markers by using the medial axis transformation. The two dimensional information from the image stick figures was compared with the projection of the three dimensional information of the model onto the relevant plane. A high degree of correlation was noted between the rotational patterns of the model and image stick figures.<>
{"title":"Lower limb kinematics of human walking with the medial axis transformation","authors":"A. G. Bharatkumar, K. E. Daigle, Marcus G. Pandy, Q. Cai, Jake K. Aggarwal","doi":"10.1109/MNRAO.1994.346252","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346252","url":null,"abstract":"The paper describes a simple model for free-speed human walking and compares ordinary images of a walking person to the model. Three dimensional kinematic data were obtained from subjects walking with markers over the joints of their limbs. The average of these data was used to derive a model stick figure of the lower limbs, based on the average anthropometric data of the population. Stick figures were obtained from ordinary images of persons dressed in tight fitting clothes without any markers by using the medial axis transformation. The two dimensional information from the image stick figures was compared with the projection of the three dimensional information of the model onto the relevant plane. A high degree of correlation was noted between the rotational patterns of the model and image stick figures.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"80 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123142168","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346238
S. Seitz, C. Dyer
Real cyclic motions tend not to be perfectly even, i.e., the period varies slightly from one cycle to the next, because of physically important changes in the scene. A generalization of period is defined for cyclic motions that makes periodic variation explicit. This representation, called the period trace, is compact and purely temporal, describing the evolution of an object or scene without reference to spatial quantities such as position or velocity. By delimiting cycles and identifying correspondences across cycles, the period trace provides a means of temporally registering a cyclic motion. In addition, several purely temporal motion features are derived, relating to the nature and location of irregularities. Results are presented using real image sequences and applications to athletic and medical motion analysis are discussed.<>
{"title":"Detecting irregularities in cyclic motion","authors":"S. Seitz, C. Dyer","doi":"10.1109/MNRAO.1994.346238","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346238","url":null,"abstract":"Real cyclic motions tend not to be perfectly even, i.e., the period varies slightly from one cycle to the next, because of physically important changes in the scene. A generalization of period is defined for cyclic motions that makes periodic variation explicit. This representation, called the period trace, is compact and purely temporal, describing the evolution of an object or scene without reference to spatial quantities such as position or velocity. By delimiting cycles and identifying correspondences across cycles, the period trace provides a means of temporally registering a cyclic motion. In addition, several purely temporal motion features are derived, relating to the nature and location of irregularities. Results are presented using real image sequences and applications to athletic and medical motion analysis are discussed.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114732556","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346251
R. Polana, Randal NelsonDepartment
The recognition of human movements such as walking, running or climbing has been approached previously by tracking a number of feature points and either classifying the trajectories directly or matching them with a high-level model of the movement. A major difficulty with these methods is acquiring and trading the requisite feature points, which are generally specific joints such as knees or angles. This requires previous recognition and/or part segmentation of the actor. We show that the recognition of walking or any repetitive motion activity can be accomplished on the basis of bottom up processing, which does not require the prior identification of specific parts, or classification of the actor. In particular, we demonstrate that repetitive motion is such a strong cue, that the moving actor can be segmented, normalized spatially and temporally, and recognized by matching against a spatiotemporal template of motion features. We have implemented a real-time system that can recognize and classify repetitive motion activities in normal gray-scale image sequences.<>
{"title":"Low level recognition of human motion (or how to get your man without finding his body parts)","authors":"R. Polana, Randal NelsonDepartment","doi":"10.1109/MNRAO.1994.346251","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346251","url":null,"abstract":"The recognition of human movements such as walking, running or climbing has been approached previously by tracking a number of feature points and either classifying the trajectories directly or matching them with a high-level model of the movement. A major difficulty with these methods is acquiring and trading the requisite feature points, which are generally specific joints such as knees or angles. This requires previous recognition and/or part segmentation of the actor. We show that the recognition of walking or any repetitive motion activity can be accomplished on the basis of bottom up processing, which does not require the prior identification of specific parts, or classification of the actor. In particular, we demonstrate that repetitive motion is such a strong cue, that the moving actor can be segmented, normalized spatially and temporally, and recognized by matching against a spatiotemporal template of motion features. We have implemented a real-time system that can recognize and classify repetitive motion activities in normal gray-scale image sequences.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"1998 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128238104","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346232
R. J. Black, Allan D. Jepson
This paper presents a new model for optical flow based on the motion of planar regions plus local deformations. The approach exploits brightness information to organize and constrain the interpretation of the motion by using segmented regions of piecewise smooth brightness to hypothesize planar regions in the scene. Parametric flow models are fitted to these regions an a two step process which first computes a coarse fit and then refines it using a generalization of the standard area-based regression approaches. Since the assumption of planarity is likely to be violated, we allow local deformations from the planar assumption. This parametric+deformation model exploits the strong constraints of parametric approaches while retaining the adaptive nature of regularization approaches.<>
{"title":"Estimating multiple independent motions in segmented images using parametric models with local deformations","authors":"R. J. Black, Allan D. Jepson","doi":"10.1109/MNRAO.1994.346232","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346232","url":null,"abstract":"This paper presents a new model for optical flow based on the motion of planar regions plus local deformations. The approach exploits brightness information to organize and constrain the interpretation of the motion by using segmented regions of piecewise smooth brightness to hypothesize planar regions in the scene. Parametric flow models are fitted to these regions an a two step process which first computes a coarse fit and then refines it using a generalization of the standard area-based regression approaches. Since the assumption of planarity is likely to be violated, we allow local deformations from the planar assumption. This parametric+deformation model exploits the strong constraints of parametric approaches while retaining the adaptive nature of regularization approaches.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"64 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133869988","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 : 1994-11-11DOI: 10.1109/MNRAO.1994.346242
H. Delingette
Simplex meshes are simply-connected meshes that are topologically the dual of triangulations. We have introduced a simplex mesh representation for recognizing partially-occluded smooth objects. In this paper, we present a physically-based approach for recovering 3D objects, based on the geometry of simplex meshes. Elastic behavior is modelled by local stabilizing functionals controlling the mean curvature through the simplex angle extracted at each vertex. Those functionals are viewpoint-invariant, intrinsic and scale-sensitive. Unlike deformable surfaces defined on regular grids, simplex meshes are highly adaptive structures, and we have developed a refinement process for increasing the mesh resolution at highly curved or inaccurate parts. End contours are created in a semi-automatic way. Finally, operations for connecting simplex meshes are performed to recover complex models from parts of simpler shapes.<>
{"title":"Adaptive and deformable models based on simplex meshes","authors":"H. Delingette","doi":"10.1109/MNRAO.1994.346242","DOIUrl":"https://doi.org/10.1109/MNRAO.1994.346242","url":null,"abstract":"Simplex meshes are simply-connected meshes that are topologically the dual of triangulations. We have introduced a simplex mesh representation for recognizing partially-occluded smooth objects. In this paper, we present a physically-based approach for recovering 3D objects, based on the geometry of simplex meshes. Elastic behavior is modelled by local stabilizing functionals controlling the mean curvature through the simplex angle extracted at each vertex. Those functionals are viewpoint-invariant, intrinsic and scale-sensitive. Unlike deformable surfaces defined on regular grids, simplex meshes are highly adaptive structures, and we have developed a refinement process for increasing the mesh resolution at highly curved or inaccurate parts. End contours are created in a semi-automatic way. Finally, operations for connecting simplex meshes are performed to recover complex models from parts of simpler shapes.<<ETX>>","PeriodicalId":336218,"journal":{"name":"Proceedings of 1994 IEEE Workshop on Motion of Non-rigid and Articulated Objects","volume":"237 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120948769","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：