Filters








927 Hits in 4.0 sec

Articulated mesh animation from multi-view silhouettes

Daniel Vlasic, Ilya Baran, Wojciech Matusik, Jovan Popović
2008 ACM Transactions on Graphics  
In this work, we demonstrate a practical software system for capturing such details from multi-view video recordings.  ...  The output mesh animation is enhanced with the details observed in the image silhouettes.  ...  We thank Peter Sand for helping us set up a multi-view studio to acquire and compute clean silhouette data. We are grateful to Tom Buehler for editing our video and to Emily Whiting for narrating it.  ... 
doi:10.1145/1360612.1360696 fatcat:eu6s2gfncrdcpnj7dhvq7s7vj4

Articulated mesh animation from multi-view silhouettes

Daniel Vlasic, Ilya Baran, Wojciech Matusik, Jovan Popović
2008 ACM SIGGRAPH 2008 papers on - SIGGRAPH '08  
In this work, we demonstrate a practical software system for capturing such details from multi-view video recordings.  ...  The output mesh animation is enhanced with the details observed in the image silhouettes.  ...  We thank Peter Sand for helping us set up a multi-view studio to acquire and compute clean silhouette data. We are grateful to Tom Buehler for editing our video and to Emily Whiting for narrating it.  ... 
doi:10.1145/1399504.1360696 fatcat:fb2mstrssng2dmma52caw6tx7a

3D Bird Reconstruction: a Dataset, Model, and Shape Recovery from a Single View [article]

Marc Badger, Yufu Wang, Adarsh Modh, Ammon Perkes, Nikos Kolotouros, Bernd G. Pfrommer, Marc F. Schmidt, Kostas Daniilidis
2020 arXiv   pre-print
Finally, we provide extensive multi-view keypoint and mask annotations collected from a group of 15 social birds housed together in an outdoor aviary.  ...  We then introduce a pipeline and experiments for keypoint, mask, pose, and shape regression that recovers accurate avian postures from single views.  ...  After collecting keypoint and silhouette ground truth from multiple views, we fit our avian mesh model using a multi-view optimization-based approach to learn a shape space and pose priors.  ... 
arXiv:2008.06133v1 fatcat:slgmmnx5yvcshmfy2rz42wj3ai

Lions and Tigers and Bears: Capturing Non-rigid, 3D, Articulated Shape from Images

Silvia Zuffi, Angjoo Kanazawa, Michael J. Black
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Figure 1 : Animals from Images. We generate 3D textured articulated models of animals from images.  ...  To make this possible, we use a strong prior model of articulated animal shape that we fit to the image data.  ...  Classical multi-view stereo and structure from motion (SfM) methods assume all images are captured with the same camera and that the camera moves while the scene is static.  ... 
doi:10.1109/cvpr.2018.00416 dblp:conf/cvpr/ZuffiKB18 fatcat:3mqcjyaasfbc7dhs4hxropip7e

Human Shape Estimation in a Multi-Camera Studio

J Starck, A Hilton, J Illingworth
2001 Procedings of the British Machine Vision Conference 2001  
We adopt a seamless articulated mesh model and introduce a novel shape matching technique to automatically transform the projected shape of the model to match multiple captured image silhouettes.  ...  Our approach treats the projected mesh as a deformable model and constrains the model to follow a smooth shape transformation to match each image silhouette.  ...  This paper extends current approaches to recovering animated models of people from image silhouettes [8] to both arbitrary pose and multiple, arbitrarily positioned cameras in a multi-camera studio.  ... 
doi:10.5244/c.15.59 dblp:conf/bmvc/StarckHI01 fatcat:dptqtb7kjnfxxndyt6dbyiiq4u

Motion capture using joint skeleton tracking and surface estimation

Juergen Gall, Carsten Stoll, Edilson de Aguiar, Christian Theobalt, Bodo Rosenhahn, Hans-Peter Seidel
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
This paper proposes a method for capturing the performance of a human or an animal from a multi-view video sequence.  ...  Given an articulated template model and silhouettes from a multi-view image sequence, our approach recovers not only the movement of the skeleton, but also the possibly non-rigid temporal deformation of  ...  In contrast, our approach captures both skeletal motion as well as an accurately deforming surface of an animal or human by fitting a body model to multi-view image data.  ... 
doi:10.1109/cvpr.2009.5206755 dblp:conf/cvpr/GallSATRS09 fatcat:5ktomw6lajfddhcpt7dj6ato64

Motion capture using joint skeleton tracking and surface estimation

J. Gall, C. Stoll, E. de Aguiar, C. Theobalt, B. Rosenhahn, H.-P. Seidel
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
This paper proposes a method for capturing the performance of a human or an animal from a multi-view video sequence.  ...  Given an articulated template model and silhouettes from a multi-view image sequence, our approach recovers not only the movement of the skeleton, but also the possibly non-rigid temporal deformation of  ...  In contrast, our approach captures both skeletal motion as well as an accurately deforming surface of an animal or human by fitting a body model to multi-view image data.  ... 
doi:10.1109/cvprw.2009.5206755 fatcat:rqpj3qcwvzhm5phqkrcvhtrvjm

LASR: Learning Articulated Shape Reconstruction from a Monocular Video [article]

Gengshan Yang, Deqing Sun, Varun Jampani, Daniel Vlasic, Forrester Cole, Huiwen Chang, Deva Ramanan, William T. Freeman, Ce Liu
2021 arXiv   pre-print
Without using a category-specific shape template, our method faithfully reconstructs nonrigid 3D structures from videos of human, animals, and objects of unknown classes.  ...  It adopts an analysis-by-synthesis strategy that forward-renders object silhouette, optical flow, and pixel values to compare with video observations, which generates gradients to adjust the camera, shape  ...  Category mesh reconstruction: A recent trend is to reduce supervision from 3D or multi-view capturing to 2D annotations, such as keypoints and object silhouettes [18, 23, 31] .  ... 
arXiv:2105.02976v1 fatcat:oostakof65ah7epconiy63y24y

Cage-Based Tracking for Performance Animation [chapter]

Yann Savoye, Jean-Sébastien Franco
2011 Lecture Notes in Computer Science  
Our results show that estimated parameters allow a sufficient silhouette-consistent generation of the enclosed mesh under sparse frame-to-frame animation constraints and large deformation.  ...  Highly-detailed performance animations are easier to obtain using existing multiple views platforms, markerless capture and 3D laser scanner.  ...  Moreover Vlasic et al. present in [4] a framework for articulated mesh animation from multi-view silhouettes.  ... 
doi:10.1007/978-3-642-19318-7_47 fatcat:worend3tprevlbv5g5rbcmu4gi

Stylizing Motion with Drawings [article]

Yin Li, Michael Gleicher, Ying-Qing Xu, Heung-Yeung Shum
2003 Proceedings of the 2004 ACM SIGGRAPH/Eurographics symposium on Computer animation - SCA '04  
We allow a traditional animator to modify frames in the rendered animation by redrawing the key features such as silhouette curves. These changes are then integrated into the animation.  ...  To perform this integration, we divide the changes into those that can be made by altering the skeletal animation, and those that must be made by altering the character's mesh geometry.  ...  Acknowledgement We would like to thank animator Yuanyuan Su for providing the drawings used in the examples.  ... 
doi:10.2312/sca03/309-319 fatcat:gago2v552rbu3o4sqnsze4qd5i

Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop [article]

Benjamin Biggs, Oliver Boyne, James Charles, Andrew Fitzgibbon, Roberto Cipolla
2020 arXiv   pre-print
We introduce an automatic, end-to-end method for recovering the 3D pose and shape of dogs from monocular internet images.  ...  We demonstrate results on the Stanford Dog dataset, an 'in the wild' dataset of 20,580 dog images for which we have collected 2D joint and silhouette annotations to split for training and evaluation.  ...  SMALR [33] then demonstrated fitting to broader animal categories by incorporating multi-view constraints from video sequences. Biggs et al.  ... 
arXiv:2007.11110v1 fatcat:4uagcg56hbdmrp7xncnkia2g7i

Marker-Less 3D Feature Tracking for Mesh-Based Human Motion Capture [chapter]

Edilson de Aguiar, Christian Theobalt, Carsten Stoll, Hans-Peter Seidel
2007 Lecture Notes in Computer Science  
Our method can therefore be used to directly animate high quality geometry models from unaltered video data which opens the door to new applications in motion capture, 3D Video and computer animation.  ...  By using the paths of the 3D points as constraints in a fast mesh deformation approach, we can directly animate a static human body scan such that it performs the same motion as the captured subject.  ...  Alternatively, shape-from-silhouette algorithms [7] , multi-view stereo approaches [14] , or methods combining silhouette and stereo constraints [8] can be used to reconstruct dynamic scene geometry  ... 
doi:10.1007/978-3-540-75703-0_1 fatcat:lmlfdvvyubhffauxwfqmahpml4

hSMAL: Detailed Horse Shape and Pose Reconstruction for Motion Pattern Recognition [article]

Ci Li, Nima Ghorbani, Sofia Broomé, Maheen Rashid, Michael J. Black, Elin Hernlund, Hedvig Kjellström, Silvia Zuffi
2021 arXiv   pre-print
Our approach is based on the SMAL model, a 3D articulated statistical model of animal shape.  ...  We test the accuracy of our hSMAL model in reconstructing a horse from 3D mocap data and images.  ...  SMAL defines a 3D articulated mesh, where the animal shape is generated with a low-dimensional linear model as a vertex deformation from a reference template.  ... 
arXiv:2106.10102v1 fatcat:p27fwlvx3jgxvabva6qtdhofwu

Iterative cage-based registration from multi-view silhouettes

Yann Savoye
2013 Proceedings of the 10th European Conference on Visual Media Production - CVMP '13  
Consistent dynamic shape capture from casual videos is a fundamental task at the cross-fertilization of Computer Vision and Computer Graphics.  ...  In this paper, we propose a cage-based technique to register non-rigid observed shapes using a meaningful animator-friendly non-rigid reduced embedding.  ...  the input multi-view silhouettes.  ... 
doi:10.1145/2534008.2534010 dblp:conf/cvmp/Savoye13 fatcat:l3xg24wrafe7fg5bqogr65m25u

4D Mesh Reconstruction from Time-Varying Voxelized Geometry through ARAP Tracking [article]

Ludovic Blache, Mathieu Desbrun, Celine Loscos, Laurent Lucas
2015 Eurographics State of the Art Reports  
Multi-view reconstruction studios use a multiple camera set to turn an actor's performance into a time series of visual hulls.  ...  We propose an automated tracking approach to convert the raw input sequence into a single, animated mesh.  ...  Model-based multi-view reconstruction approaches use a template model representing an actor, typically, an articulated mesh of a generic human body [VBMP08, GSdA * 09].  ... 
doi:10.2312/egsh.20151002 fatcat:zenuwfwswbeizajsqoic6vyor4
« Previous Showing results 1 — 15 out of 927 results