期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2014
卷号：2
期号：7
出版社：S&S Publications
摘要：Face recognition systems are becoming popular in most of the applications ranging from gaming tosurveillance. Conventionally, 2D cameras are used in face recognition applications and hence 2D face recognitionalgorithms are employed. However, it has become very easy to obtain 3D camera at cheaper cost, now days. It isinteresting to see how 3D camera information can be used to overcome the challenges faced by conventional 2D facealgorithms. Recently Creative Labs launched Senz3D camera which outputs the RGB and depth images withmisalignment in object present in them. To overcome this, we proposed two algorithms in this paper. It is found thatthese algorithms have improved the recognition performance significantly. We also explored the various featurerepresentations for the fusion of RGB and depth face ROIs. The fusion of RGB and depth ROIs representation isachieved at matching score level with nearest neighbor classifier. It is observed that among all combinations of featurerepresentations for RGB and depth images, two methods, HOG (RGB) +LDA (Depth) and LBP (RGB) +LDA (Depth),have performed better for fusion system.Additionally, the fusion can play more critical role in the situations where lightis not present in vicinity by adaptively increasing the fusion weight associated with depth information.
关键词：2D Face; 3D Face; RGB-D camera; Face Recognition; Fusion; Depth image.
