Institute Homepage
Institute Homepage DE Sign In

Back

Perceiving Systems Conference Paper 1998

Motion feature detection using steerable flow fields

pdf
Thumb ticker sm headshot2021
Perceiving Systems
Michael Black
Director
no image
Perceiving Systems
Allan Jepson
Bildschirmfoto 2013 01 14 um 09.23.21

The estimation and detection of occlusion boundaries and moving bars are important and challenging problems in image sequence analysis. Here, we model such motion features as linear combinations of steerable basis flow fields. These models constrain the interpretation of image motion, and are used in the same way as translational or affine motion models. We estimate the subspace coefficients of the motion feature models directly from spatiotemporal image derivatives using a robust regression method. From the subspace coefficients we detect the presence of a motion feature and solve for the orientation of the feature and the relative velocities of the surfaces. Our method does not require the prior computation of optical flow and recovers accurate estimates of orientation and velocity.

Author(s): Fleet, D. J. and Black, M. J. and Jepson, A. D.
Links:
Book Title: IEEE Conf. on Computer Vision and Pattern Recognition, CVPR-98
Pages: 274-281
Year: 1998
Publisher: IEEE
Bibtex Type: Conference Paper (inproceedings)
Address: Santa Barbara, CA
Electronic Archiving: grant_archive

BibTex 

@inproceedings{Fleet:IEEE:1998,
  title = {Motion feature detection using steerable flow fields},
  booktitle = {IEEE Conf. on Computer Vision and Pattern Recognition, CVPR-98},
  abstract = {The estimation and detection of occlusion boundaries and moving bars are important and challenging problems in image sequence analysis. Here, we model such motion features as linear combinations of steerable basis flow fields. These models constrain the interpretation of image motion, and are used in the same way as translational or affine motion models. We estimate the subspace coefficients of the motion feature models directly from spatiotemporal image derivatives using a robust regression method. From the subspace coefficients we detect the presence of a motion feature and solve for the orientation of the feature and the relative velocities of the surfaces. Our method does not require the prior computation of optical flow and recovers accurate estimates of orientation and velocity.},
  pages = {274-281},
  publisher = {IEEE},
  address = {Santa Barbara, CA},
  year = {1998},
  slug = {fleet-ieee-1998},
  author = {Fleet, D. J. and Black, M. J. and Jepson, A. D.}
}