Author:
Séverine Dubuisson
Affiliation:
Laboratoire d’Informatique de Paris 6 (LIP6/UPMC), France
Keyword(s):
Fast histogram computation, Integral histogram.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Early Vision and Image Representation
;
Feature Extraction
;
Features Extraction
;
Image and Video Analysis
;
Informatics in Control, Automation and Robotics
;
Signal Processing, Sensors, Systems Modeling and Control
Abstract:
In this paper we present a new method for fast histogram computing. Based on the known tree-representation histogram of a region, also called reference histogram,, we want to compute the one of another region. The idea consists in computing the spatial differences between these two regions and encode it to update the histogram. We never need to store complete histograms, except the reference image one (as a preprocessing step). We compare our approach with the well-known integral histogram, and obtain better results in terms of processing time while reducing the memory footprint. We show theoretically and with experimental results the superiority of our approach in many cases. Finally, we demonstrate the advantage of this method on a visual tracking application using a particle filter by improving its time computing.