Repository logo
Log in(current)
Repository logoMIT Open ScholarshipDSpace@MIT
  1. Home
  2. Computer Science and Artificial Intelligence Lab (CSAIL)
  3. Artificial Intelligence Lab Publications
  4. AI Memos (1959 - 2004)
  5. Estimation of Discontinuous Displacement Vector Fields with the Minimum Description Length Criterion

Estimation of Discontinuous Displacement Vector Fields with the Minimum Description Length Criterion

Thumbnail Image
Download
Name

AIM-1265.pdf

Size

1.67 MB

Format

Adobe PDF

Checksum (MD5)

bdac7a51aa8203c443e8cf0d3ca0f415

Thumbnail Image
Download
Name

AIM-1265.ps

Size

4.2 MB

Format

PostScript

Checksum (MD5)

78281b280f60e3f74160242d518bbd74

Author(s)
Dengler, Joachim
Date Issued
October 1, 1990
Series/Report no.
AIM-1265
Abstract
A new noniterative approach to determine displacement vector fields with discontinuities is described. In order to overcome the limitations of current methods, the problem is regarded as a general modelling problem. Starting from a family of regularized estimates, by measuring the difference in description length the compatibility between different levels of regularization is determined. This gives local but noisy evidence of possible model boundaries at multiple scales. With the two constraints of continous lines of discontinuities and the spatial coincidence assumption consistent boundary evidence is found. Based on this combined evidence the model is updated, now describing homogeneous regions with sharp discontinuities.
Subjects
correspondence problem
pyramids
motion stereo
sdiscontinuities
Persistent DSpace Link
http://hdl.handle.net/1721.1/5993
Repository logo
PrivacyPermissionsAccessibilityContact us
Repository logo
Notify us about copyright concerns.