ProjectGeodesicNeighborhoods

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

Geodesic pixel neighborhoods for multi-class image segmentation

Geodesic pixel neighborhoods for multi-class image segmentation

Scientific Director: Slobodan Ilic

Contact Person(s): Vladimir Haltakov

In industrial collaboration with:
BMW Group

News

  • 01.09.2014 - Paper presented on the British Machine Vision Conference in Nottingham.

Description

The problem of multi-class image segmentation is traditionally addressed either with a Conditional Random Field model or by directly classifying each pixel. In this paper we introduce a new classification framework built around the concept of pixel neighborhoods. A standard unary classifier is used to recognize each pixel based on features computed from the image and then a second classifier is trained on the predictions from the first one summarized over the neighborhood of each pixel by a new histogram feature. We define a local and a global pixel neighborhoods which adapt to the image structure by making use of the geodesic distance defined over image intensities.

We evaluate our model on three challenging datasets and show that our model is able to capture both local and global context relations. We compare our method to two strongly related, well known methods and show increased performance.

Related work

2014
V. Haltakov , C. Unger, S. Ilic
Geodesic pixel neighborhoods for multi-class image segmentation
British Machine Vision Conference, 1.-5. September 2014 (bib)

2016
V. Haltakov , C. Unger, S. Ilic
Geodesic pixel neighborhoods for 2D and 3D scene understanding.
Computer Vision and Image Understanding, Volume 148, July 2016, Pages 164-180 (bib)





Edit | Attach | Refresh | Diffs | More | Revision r1.2 - 05 Feb 2017 - 15:50 - VladimirHaltakov

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif