The Community for Technology Leaders
2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (2012)
Providence, RI USA
June 16, 2012 to June 21, 2012
ISSN: 2160-7508
ISBN: 978-1-4673-1611-8
pp: 58-63
Mayank Rana , Courant Institute of Mathematical Sciences, New York University, New York, USA
Graham Taylor , Courant Institute of Mathematical Sciences, New York University, New York, USA
Ian Spiro , Courant Institute of Mathematical Sciences, New York University, New York, USA
Christoph Bregler , Courant Institute of Mathematical Sciences, New York University, New York, USA
ABSTRACT
This paper demonstrates how 3D skeletal reconstruction can be performed by using a pose-sensitive embedding technique applied to multi-view video recordings. We apply our approach to challenging low-resolution video sequences. Usually skeletal reconstruction can be only achieved with many calibrated high-resolution cameras, and only blob detection can be achieved with such low-resolution imagery. We show that with this embedding technique (a metric learning technique using a deep convolutional architecture), we achieve very good 3D skeletal reconstruction on low-resolution outdoor scenes with many challenges.
INDEX TERMS
image reconstruction, image resolution, image sequences, natural scenes, pose estimation, video recording, video signal processing
CITATION

M. Rana, G. Taylor, I. Spiro and C. Bregler, "3D skeletal reconstruction from low-resolution multi-view images," 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops(CVPRW), Providence, RI USA, 2012, pp. 58-63.
doi:10.1109/CVPRW.2012.6239238
96 ms
(Ver 3.3 (11022016))