Authors:
M.-A. Bauda
1
;
S. Chambon
2
;
P. Gurdjos
2
and
V. Charvillat
2
Affiliations:
1
University of Toulouse and imajing sas, France
;
2
University of Toulouse, France
Keyword(s):
Image Segmentation, Superpixel, Planar Hypothesis.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Applications and Services
;
Color and Texture Analyses
;
Computer Vision, Visualization and Computer Graphics
;
Geometry and Modeling
;
Image and Video Analysis
;
Image Formation and Preprocessing
;
Image Formation, Acquisition Devices and Sensors
;
Image-Based Modeling
;
Mobile Imaging
;
Pattern Recognition
;
Segmentation and Grouping
;
Shape Representation and Matching
;
Software Engineering
Abstract:
Superpixel segmentation is widely used in the preprocessing step of many applications. Most of existing methods are based on a photometric criterion combined to the position of the pixels. In the same way as the Simple Linear Iterative Clustering (SLIC) method, based on k-means segmentation, a new algorithm is introduced. The main contribution lies on the definition of a new distance for the construction of the superpixels. This distance takes into account both the surface normals and a similarity measure between pixels that are located on the same planar surface. We show that our approach improves over-segmentation, like SLIC, i.e. the proposed method is able to segment properly planar surfaces.