8,086 Hits in 5.1 sec

Home 3D body scans from noisy image and range data

Alexander Weiss, David Hirshberg, Michael J. Black
2011 2011 International Conference on Computer Vision  
We present a method for human shape reconstruction from noisy monocular image and range data using a single inexpensive commodity sensor.  ...  The 3D shape of the human body is useful for applications in fitness, games and apparel. Accurate body scanners, however, are expensive, limiting the availability of 3D body models.  ...  First they provide a height-and weight-constrained subspace of body shape variation to constrain the problem.  ... 
doi:10.1109/iccv.2011.6126465 dblp:conf/iccv/WeissHB11 fatcat:iu4ojrl7v5cbjac4xsf5w3ld7y

Constraint-fusion for interpretation of articulated objects

Hel-Or, Werman
1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition CVPR-94  
This paper presents a method for interpretation of modeled objects that is general enough to cover articulated and other types of constrained m o dels.  ...  The measurements taken on these points are noisy. 2 Formal Description of the Problem A constrained model M of a 3D object consists of a set of rigid components M = fC i g i=1 n : Each component C i has  ...  A set of points forming a component is rigid but the collection of components are not rigid.  ... 
doi:10.1109/cvpr.1994.323808 dblp:conf/cvpr/Hel-OrW94 fatcat:g3ujip2rdre2laetp4gotn7dua

Robust sequence alignment for actor–object interaction recognition: Discovering actor–object states

Roman Filipovych, Eraldo Ribeiro
2011 Computer Vision and Image Understanding  
We use the actor-object state concept to propose an approach in which human-object interactions are represented by a combination of image patches and velocity information extracted along tracked body-point  ...  ., shape) of the interacting person are constrained by the target object.  ...  Notice how points outside the central pattern of interest, as well as noisy points, are involved in the DTW's warping function.  ... 
doi:10.1016/j.cviu.2010.11.012 fatcat:v7mqjphyerevncresradquytyi

Line-based extrinsic calibration of range and image sensors

Peyman Moghadam, Michael Bosse, Robert Zlot
2013 2013 IEEE International Conference on Robotics and Automation  
Fig. 1: Overview of the proposed method: lines are extracted from (a) an image and (b) from a point cloud of the same scene.  ...  The majority of existing extrinsic calibration techniques require one or multiple planar calibration patterns (such as checkerboards) to be observed simultaneously from the range and imaging sensors.  ...  ., they intersect at least on one common point), the optimization will be weakly constrained.  ... 
doi:10.1109/icra.2013.6631095 dblp:conf/icra/MoghadamBZ13 fatcat:dgq4szdkxzefrfnh4emtxpxp7y

Dynamics Based 3D Skeletal Hand Tracking [article]

Stan Melax, Leonid Keselman, Sterling Orsten
2017 arXiv   pre-print
This research explores a new approach to tracking hands, or any articulated model, by using an augmented rigid body simulation.  ...  Based on a depth sensor's samples, the system generates constraints that limit motion orthogonal to the rigid body model's surface.  ...  Explanation of Rigid Body Dynamics Rigid body physics systems simulate the interaction of rigid bodies [6, 41] .  ... 
arXiv:1705.07640v1 fatcat:wzpkyfjqijakpfm3wxi3th4ktm

Free-form mesh tracking: A patch-based approach

Cedric Cagniart, Edmond Boyer, Slobodan Ilic
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
This strategy ensures robustness by providing natural integration domains over the surface for noisy data, while enabling to express simple patch-level rigidity constraints.  ...  We also performed quantitative and qualitative evaluations in multicameras and monocular environments, and with different types of data including 2D correspondences and 3D point clouds.  ...  Note on the two first images that in absence of matches, the rigidity constraints make the mesh return locally to its rest pose (flat cloth). .  ... 
doi:10.1109/cvpr.2010.5539814 dblp:conf/cvpr/CagniartBI10 fatcat:ubbkiibvefcb3psholio3lhwna

Animated statues

Jonathan Starck, Gordon Collins, Raymond Smith, Adrian Hilton, John Illingworth
2003 Machine Vision and Applications  
A set of manually defined feature points are used to define body and facial pose, and a novel shape constrained matching algorithm is presented to deform the control model to match the scanned shape.  ...  We introduce a model-based approach to animating a scanned data-set by matching a generic humanoid control model to the surface data.  ...  Model to Surface Matching The shape constrained deformable model is matched to the body data in order to deform the model to fit the shape of the body.  ... 
doi:10.1007/s00138-002-0080-3 fatcat:klduhezosbgktevqu5oclyhena

Depth Super Resolution by Rigid Body Self-Similarity in 3D

Michael Hornacek, Christoph Rhemann, Margrit Gelautz, Carsten Rother
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
6 DoF rigid body motion in 3D.  ...  Rather than reason in terms of patches of 2D pixels as others have before us, our key contribution is to proceed by reasoning in terms of patches of 3D points, with matched patch pairs related by a respective  ...  and which we match with respect to 3D point similarity over 6 DoF rigid body motions in 3D.  ... 
doi:10.1109/cvpr.2013.149 dblp:conf/cvpr/HornacekRGR13 fatcat:e5iffmrloba4jidtv6w2so4ipe

Non-rigid Point Set Registration with Global-Local Topology Preservation

Song Ge, Guoliang Fan, Meng Ding
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
We propose a new point set registration method, Global-Local Topology Preservation (GLTP), which can cope with complex non-rigid transformations including highly articulated deformation.  ...  We have derived the EM algorithm for the ML optimization constrained with both CPD and LLE terms, leading to the new GLTP algorithm.  ...  The two point sets have different body shapes (non-rigid local deformations) and poses (global articulation).  ... 
doi:10.1109/cvprw.2014.45 dblp:conf/cvpr/GeFD14 fatcat:kmqxisiujfauxabs66dfg4pcly

Globally optimal solution to exploit rigidity when recovering structure from motion under occlusion

Pedro M. Q. Aguiar, Joao M. F. Xavier, Marko Stosic
2008 2008 15th IEEE International Conference on Image Processing  
Widely used SVD-based matrix factorization approaches to the recovery of 3D rigid structure from motion (SFM), require a set of feature points to be visible in a set of images.  ...  When there is occlusion, several feature points disappear, the observation matrix misses some entries, there is not equivalent to the SVD, and only suboptimal solutions have been proposed to exploit rigidity  ...  INTRODUCTION A matrix collecting trajectories of feature points of a rigid body is rank deficient in a noiseless scenario.  ... 
doi:10.1109/icip.2008.4711725 dblp:conf/icip/AguiarXS08 fatcat:dyegooifubdklbgrayubar2bwe

Unsupervised Learning of Consensus Maximization for 3D Vision Problems

Thomas Probst, Danda Pani Paudel, Ajad Chhatkuli, Luc Van Gool
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Using our method, we solve a diverse set of 3D vision problems, including 3D-3D matching, non-rigid 3D shape matching with piece-wise rigidity and image-toimage matching.  ...  Rigid Body Transformation We consider correspondences between two point clouds that differ by a 3D rigid body transformation.  ...  Red points in the last two images are the SfM point cloud superimposed on the fitted mesh model. Figure 7 . 7 Rigid body transformation.  ... 
doi:10.1109/cvpr.2019.00102 dblp:conf/cvpr/ProbstPCG19 fatcat:dq56oeult5cpvmfvclfalk72jy

Axially Symmetric 3D Pots Configuration System Using Axis of Symmetry and Break Curve

Kilho Son, Eduardo B. Almeida, David B. Cooper
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
The estimation exploits both local and semi-global geometric structure, thus making it a fundamental problem of geometry estimation from noisy fragments in computer vision and pattern recognition.  ...  In 3D, each fragment has 6 degrees of freedom (DOF) for rigid body motion.  ...  This process is very robust to noisy data. Iterative closest point (ICP) algorithm is used to compute a rigid transformation required for alignment.  ... 
doi:10.1109/cvpr.2013.40 dblp:conf/cvpr/SonAC13 fatcat:wnwew3l7vrb6lhucvew4qjg5im

Non-rigid structure from locally-rigid motion

Jonathan Taylor, Allan D. Jepson, Kiriakos N. Kutulakos
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
We introduce locally-rigid motion, a general framework for solving the M -point, N -view structure-from-motion problem for unknown bodies deforming under orthography.  ...  The key idea is to first solve many local 3-point, N -view rigid problems independently, providing a "soup" of specific, plausibly rigid, 3D triangles.  ...  ., that all four scene points belong to the same rigid or non-rigid body.  ... 
doi:10.1109/cvpr.2010.5540002 dblp:conf/cvpr/TaylorJK10 fatcat:euvdenjj7nggfoagfduuiymcvi

Nonrigid Motion Analysis: Articulated and Elastic Motion

J.K. Aggarwal, Q. Cai, W. Liao, B. Sabata
1998 Computer Vision and Image Understanding  
In articulated motion, individual rigid parts of an object move independently of one another and the motion of the whole object is nonrigid in nature.  ...  The nonrigid objects being studied may be generally categorized into three groups according to the degree of deformation of body parts: articulated, elastic, and fluid.  ...  Implicit point matching algorithms produce denser optical flow fields, but the results are usually noisy and necessitate some smoothing process.  ... 
doi:10.1006/cviu.1997.0620 fatcat:mouq7mt7yjgmrmnwvwll3gguxm

3D Tracking Of Deformable Objects With Applications To Coding And Recognition

C. Alberola-Lopez, J.R. Casar-Corredera, G. de Miguel Vela, J. Ruiz-Alzola
1996 Zenodo  
Herein we present an approach to estimate the current state of the non-rigid body based on tracking some outstanding tokens on the body surface : 2.1.  ...  -Matching Measurements To The States. Suppose that we have got a body described by N T tokens and by a FSM with N S states.  ... 
doi:10.5281/zenodo.36105 fatcat:jbxqv4skrzhmjaabsy2cbb2z4y
« Previous Showing results 1 — 15 out of 8,086 results