A Learning Approach for Adaptive Image Segmentation

Vincent Martin 1, * Nicolas Maillot 1 Monique Thonnat 1
* Corresponding author
Abstract : As mentioned in many papers, a lot of key parameters of image segmentation algorithms are manually tuned by designers. This induces a lack of flexibility of the segmentation step in many vision systems. By a dynamic control of these parameters, results of this crucial step could be drastically improved. We propose a scheme to automatically select segmentation algorithm and tune theirs key parameters thanks to a preliminary supervised learning stage. This paper details this learning approach which is composed by three steps: (1) optimal parameters extraction, (2) algorithm selection learning, and (3) generalization of parametrization learning. The major contribution is twofold: segmentation is adapted to the image to segment, and in the same time, this scheme can be used as a generic framework, independant of any application domain.
Document type :
Conference papers
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download

https://hal.inria.fr/inria-00499629
Contributor : Vincent Martin <>
Submitted on : Sunday, July 11, 2010 - 1:09:30 PM
Last modification on : Saturday, January 27, 2018 - 1:30:44 AM
Long-term archiving on : Tuesday, October 12, 2010 - 10:09:41 AM

File

ICVS06.pdf
Publisher files allowed on an open archive

Identifiers

Collections

Citation

Vincent Martin, Nicolas Maillot, Monique Thonnat. A Learning Approach for Adaptive Image Segmentation. International Conference on Computer Vision Systems, Jan 2006, New York City, NJ, United States. pp.40, ⟨10.1109/ICVS.2006.4⟩. ⟨inria-00499629⟩

Share

Metrics

Record views

230

Files downloads

662