Filters








30,565 Hits in 13.3 sec

Finding point correspondences and determining motion of a rigid object from two weak perspective views

Chia-Hoang Lee, Thomas Huang
1990 Computer Vision Graphics and Image Processing  
(ii) What underlying motions and associated depth components of these points could account for the two images ? (ill) Can the object be recovered uniquely? This fonnulation of !  ...  Given two images of an n-point configuration which undergoes 3D rotation, translation and scaling; Our problems are (i) How can we match the corresponding points in the two images?  ...  In this paper we investigate with regard to weak perspective views how to relate the correspondence, the motion and the structure of an n-points object, given two images which might have different scales  ... 
doi:10.1016/0734-189x(90)90014-m fatcat:7caht6zzlnes5nsqspb2jbg3aa

Rigid body segmentation and shape description from dense optical flow under weak perspective

J. Weber, J. Malik
1997 IEEE Transactions on Pattern Analysis and Machine Intelligence  
The use of epipolar geometry allows for the determination of individual motion parameters for each object as well as the recovery of relative depth for each point on the object.  ...  We present an algorithm for identifying and tracking independently moving rigid objects from optical flow.  ...  Motion parameters and shape descriptions can also be obtained from correspondences between two views under weak perspective projection, modulo a relief transformation such as depth scaling [8] .  ... 
doi:10.1109/34.574794 fatcat:p45ghg7rxzd2tpurzemfru4fle

Finding Point Correspondences in Motion Sequences Preserving Affine Structure

G. Sudhir, Subhashis Banerjee, Andrew Zisserman
1993 Procedings of the British Machine Vision Conference 1993  
In this paper the problem of computing the point correspondences in 3frames of a sequence of time-varying images of a 3D object undergoing nonrigid (affine) motion is addressed.  ...  The correspondences are established only from the analysis of the unknown affine structure of the object, without, making use of any attributes of the feature points.  ...  correspondences in two weak-perspective views.  ... 
doi:10.5244/c.7.36 dblp:conf/bmvc/SudhirBZ93 fatcat:u6yjlezp2vd3dlnr3t2ri64qwy

Finding Point Correspondences in Motion Sequences Preserving Affine Structure

G. Sudhir, Subhashis Banerjee, Andrew Zisserman
1997 Computer Vision and Image Understanding  
In this paper the problem of computing the point correspondences in 3frames of a sequence of time-varying images of a 3D object undergoing nonrigid (affine) motion is addressed.  ...  The correspondences are established only from the analysis of the unknown affine structure of the object, without, making use of any attributes of the feature points.  ...  correspondences in two weak-perspective views.  ... 
doi:10.1006/cviu.1997.0545 fatcat:mbcmw5y6vjdk3ahelo5ndfx7qm

Perceiving motion and rigid structure from optic flow: A combined weak-perspective and polar-perspective approach

Mats Lind
1996 Perception & Psychophysics  
Structure-from-motion algorithms based on weak-perspective projection have many interesting properties and could serve as a basis for a model of human perception of motion and structure from motion (M&  ...  However, from a geometrical point of view, many different types of projections are conceivable.  ...  Perceiving motion and rigid structure from optic flow: A combined weak-perspective and polar-perspective approach MATS LIND Uppsala University, Uppsala, Sweden Structure-from-motion algorithms based  ... 
doi:10.3758/bf03206835 pmid:8920844 fatcat:wewjv2q25bg7xbnbmumkpysogy

Correspondence and segmentation of multiple rigid motions via epipolar geometry

Gang Xu, S. Tsuji
1996 Proceedings of 13th International Conference on Pattern Recognition  
This paper presents a new approach to matching and segmenting multiple rigid motions using epipolar constraint.  ...  Epipolar geometry underlying the motion images are rst recovered, and the epipolar equations are used to reduce search for correspondence from 2dimensional to 1-dimensional, thus resolving the \aperture  ...  Also, from a computational point of view, it has to be solved in the case of multiple motions for both correct correspondence and structure-from-motion computation.  ... 
doi:10.1109/icpr.1996.546021 dblp:conf/icpr/XuT96 fatcat:arppfqbpzzfqxmtjnsjhm6ebkq

The Interpretation of Structure from Motion

S. Ullman
1979 Proceedings of the Royal Society of London. Biological Sciences  
The interpretation of structure from motion is examined from a computational point of view.  ...  It is shown that this scheme will correctly decompose scenes containing arbitrary rigid objects In motion, recovering their 3-D structure and motion.  ...  The interpretation of structure from motion is examined from a computational point of view .  ... 
doi:10.1098/rspb.1979.0006 fatcat:lnjxyz7iabgsfowiqdt2bdxu7u

Luminance controls the perceived 3-D structure of dynamic 2-D displays

Barry J. Schwartz, George Sperling
1983 Bulletin of the Psychonomic Society  
When the cube is viewed from a close range so that linear perspective is significant (Figures lb, Ie, and ld), then it alternates perceptually from a 3-D rigid cube to a nonrigid, distorting, rubbery-looking  ...  geometric cues of form and motion that are supposed to govern the perception of rigid objects in the KDE.  ...  (e and i) Two views of the same random-wire figure, drawn without linear perspective.  ... 
doi:10.3758/bf03330007 fatcat:fhn73evmb5cxxede5y3zpw7rte

Deterministic 3D Human Pose Estimation Using Rigid Structure [chapter]

Jack Valmadre, Simon Lucey
2010 Lecture Notes in Computer Science  
This paper explores a method, first proposed by Wei and Chai [1], for estimating 3D human pose from several frames of uncalibrated 2D point correspondences containing projected body joint locations.  ...  Based on this analysis we propose a novel approach using deterministic structure from motion based on assumptions of rigidity in the body's torso.  ...  In [1] Wei and Chai claim their approach requires a minimum of five frames with 2D point correspondences (under a weak perspective assumption) for a 17 bone body containing 4 rigid triangles and 7 symmetry  ... 
doi:10.1007/978-3-642-15558-1_34 fatcat:qlwaaoelvbayxgktvgannu67am

Rigid and affine motion estimation in vision using sliding mode observers

B. Yilmaz, M. Unel, A. Sabanovic
2005 Proceedings of the IEEE International Symposium on Industrial Electronics, 2005. ISIE 2005.  
First the motion is of rigid type and constrained to a plane, and object, e.g. a mobile robot, is viewed using a stationary camera at a fixed certain distance in the normal direction of the motion plane  ...  A "sliding mode observer" based approach to estimate rigid and affine motion of an object is presented in this paper. Two different problems are considered.  ...  Thus the practical categorization of a motion into rigid or affine groups is mainly determined by the position and orientation of the camera viewing the moving object. C.  ... 
doi:10.1109/isie.2005.1528884 fatcat:vwai6vrgbncptn3kvrh24qupia

Graph Rigidity for Near-Coplanar Structure from Motion

Jack Valmadre, Ben Upcroft, Sridha Sridharan, Simon Lucey
2011 2011 International Conference on Digital Image Computing: Techniques and Applications  
Recent algorithms for monocular motion capture (MoCap) estimate weak-perspective camera matrices between images using a small subset of approximately-rigid points on the human body (i.e. the torso and  ...  In this paper, we propose an alternative solution to weak-perspective SFM based on a convex relaxation of graph rigidity.  ...  In the area of perspective geometry, Li [14] recently presented a seminal algorithm for reconstructing a 3D scene from 2D point correspondences without having to estimate camera matrices.  ... 
doi:10.1109/dicta.2011.87 dblp:conf/dicta/ValmadreUSL11 fatcat:goiebe3ihvdvncdmjx77q4auoy

Uncalibrated perspective reconstruction of deformable structures

Jing Xiao, T. Kanade
2005 Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1  
Reconstruction of 3D structures from uncalibrated image sequences has a wealthy history. Most work has been focused on rigid objects or static scenes.  ...  We develop a factorization algorithm that consists of two steps.  ...  Introduction Perspective reconstruction of 3D structure and motion from a 2D image sequence has been studied for decades.  ... 
doi:10.1109/iccv.2005.241 dblp:conf/iccv/XiaoK05 fatcat:va2lve7fanh3tbpztouyduvvwu

Modelling Nonrigid Object from Video Sequence Under Perspective Projection [chapter]

Guanghui Wang, Yantao Tian, Guoqiang Sun
2005 Lecture Notes in Computer Science  
The paper is focused on the problem of estimating 3D structure and motion of nonrigid object from a monocular video sequence.  ...  The proposed algorithm iteratively updates the shape and motion from weak perspective projection to fully perspective projection by refining the scalars corresponding to the projective depths.  ...  Under weak perspective projection, the centroid of a set of space points and that of their image points correspond to each other.  ... 
doi:10.1007/11573548_9 fatcat:f5jc3daanvefhd33pt6d76uuge

Structure and motion of nonrigid object under perspective projection

Guanghui Wang, Hung-Tat Tsui, Zhanyi Hu
2007 Pattern Recognition Letters  
The paper focuses on the problem of structure and motion of nonrigid object from image sequence under perspective projection.  ...  Both algorithms are based on the shape and motion of weak perspective projection.  ...  The work is supported by the National Natural Science Foundation of China under Grant no. 60575015 and the Hong Kong RGC Grant CUHK 4378/02E.  ... 
doi:10.1016/j.patrec.2006.09.006 fatcat:klr3pn6jvnczjjvhfivx6iirfi

Detecting group activities using rigidity of formation

Saad M. Khan, Mubarak Shah
2005 Proceedings of the 13th annual ACM international conference on Multimedia - MULTIMEDIA '05  
Our method is invariant to camera motion and does not require an a priori model or a training phase.  ...  We model the formation as a 3D polygon with each corner representing a participating entity. Tracks from the entities are treated as tracks of feature points on the 3D polygon.  ...  treated as a rigid object) then it has been shown by Tomasi and Kanade [3] that if the camera is affine (orthographic or weak-perspective) and when there is no noise, then the rank of W is 3 or lower  ... 
doi:10.1145/1101149.1101237 dblp:conf/mm/KhanS05 fatcat:q7l5s23b35b5tom4ziefbmobpq
« Previous Showing results 1 — 15 out of 30,565 results