ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper An Expression-Invariant Facial Image Retrieval using SIFT and GLBP
Cited - time in scopus Share share facebook twitter linkedin kakaostory
Authors
Sohee Park, Geonwoo Kim
Issue Date
2017-12
Citation
International Conference on Computer Science and its Applications (CSA) 2017, pp.1-6
Language
English
Type
Conference Paper
Abstract
This paper presents an expression-invariant face search method for the real environment. This method is based on Scale Invariant Feature Transform (SIFT) matching and Gabor – Local Binary Pattern (GLBP) histogram for simple and non-statistical approach. We already presented the performance of SIFT and GLBP based face search with partial occlusion. The main characteristic of this approach is to measure the similarity between two images using SIFT. In this paper, we presents the performance of our approach in the expression-invariant face search as well as occluded face images. For demonstration the superiority of our approach, we used the famous benchmark database with expressional face images. The results show that our approach achieves better performance compared to GLBP search algorithm on AR face database.
KSP Keywords
AR face database, Face image, Facial image, Image retrieval, Occluded Face, Partial Occlusion, Real environment, Search Algorithm(GSA), local binary pattern(LBP), measure the similarity, scale invariant feature transform(SIFT)