Authors:
Sinem Aslan
1
;
Ceyhun Burak Akgül
2
;
Bülent Sankur
2
and
Turhan Tunalı
3
Affiliations:
1
Ege University, Turkey
;
2
Boğaziçi University, Turkey
;
3
Izmir University of Economics, Turkey
Keyword(s):
Image Feature, Model-driven Visual Dictionary, Primitive Structures of Natural Images, Object Recognition, Image Understanding.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Features Extraction
;
Image and Video Analysis
;
Image Formation and Preprocessing
;
Image Formation, Acquisition Devices and Sensors
;
Image Generation Pipeline: Algorithms and Techniques
;
Shape Representation and Matching
Abstract:
We propose a new local image descriptor named SymPaD for image understanding. SymPaD is a probability
vector associated with a given image pixel and represents the attachment of the pixel to a previously designed
shape repertoire. As such the approach is model-driven. The SymPad descriptor is illumination and rotation
invariant, and extremely flexible on extending the repertoire with any parametrically generated geometrical
shapes and any desired additional transformation types.