Title :
A probabilistic framework for logo detection and localization in natural scene images
Author :
Roy, Anirban ; Garain, U.
Author_Institution :
Comput. Vision & Pattern Recognition Unit, Indian Stat. Inst., Kolkata, India
Abstract :
This paper presents a probabilistic approach for logo detection and localization in natural scene images. Two probability distributions are computed, one considering the features extracted from the key points located inside a region and the second refers to shape geometry defined by the key points. The barycentric co-ordinates are considered to define the shape statistics. The performance of the proposed approach has been reported on two publicly available datasets: BelgaLogos and Flickr Logos27. It is shown that statistically significant improvement is achieved over a recently proposed method.
Keywords :
feature extraction; geometry; natural scenes; object detection; shape recognition; statistical distributions; BelgaLogos; Flickr Logos27; barycentric co-ordinates; feature extraction; natural scene images; probabilistic logo detection framework; probabilistic logo localization framework; probability distribution; publicly available datasets; shape geometry; shape statistics; statistically significant improvement; Accuracy; Databases; Feature extraction; Geometry; Probabilistic logic; Shape; Visualization;
Conference_Titel :
Pattern Recognition (ICPR), 2012 21st International Conference on
Conference_Location :
Tsukuba
Print_ISBN :
978-1-4673-2216-4