System and method for automatic landmark labeling with minimal supervision
US8442330B2 · kind B2 · utility
5Cited by
12References
16Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Jul 31, 2009 |
| Grant date | May 14, 2013 |
| Priority date | — |
| Expiry date | Jan 24, 2032 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30201
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.