Title :
A quantitative metric of visual-words separability for a more discriminative visual vocabulary in an unsupervised manner
Author :
Xin Feng ; Bo Li ; Yongxin Ge ; Jiaxing Tan
Author_Institution :
Coll. of Comput. Sci. & Eng., Chongqing Univ. of Technol., Chongqing, China
Abstract :
The task of visual vocabulary construction plays an important role in the bag-of-words based pattern analysis and robotic applications. A discriminative vocabulary generation in unsupervised case is an open issue for reducing perceptual aliasing in image matching based applications. In this paper, we present a scheme to evaluate the discriminative power of each visual word quantitatively in terms of Mahalanobis separability, and a discriminative visual vocabulary is obtained through adaptively updating the poor discriminative visual words in an unsupervised manner. The effectiveness of our metric is demonstrated in the experiment of loop-closure detection under strong perceptual aliasing condition in both indoor and outdoor image sequences.
Keywords :
image matching; image sequences; vocabulary; Mahalanobis separability; bag-of-words; discriminative visual vocabulary; image matching; indoor image sequences; outdoor image sequences; pattern analysis; quantitative metric; robotic applications; unsupervised manner; visual vocabulary construction; visual words separability; Cameras; Computational modeling; Indexes; Optimization; Robots; Visualization; Vocabulary; loop closure detection; mahalanobis separability; visual vocabulary;
Conference_Titel :
Information, Communications and Signal Processing (ICICS) 2013 9th International Conference on
Conference_Location :
Tainan
Print_ISBN :
978-1-4799-0433-4
DOI :
10.1109/ICICS.2013.6782779
