Search

Currently, most face recognition methods provide a number of parameters to be optimized, leaving the selection and optimization of the right parameter set is necessary for the implementation. The choice of the right parameter set that is suitable for a rich enough class of input faces in pose and illumination variations is, however, quite difficult. We propose robust parameter estimation, using the Taguchi method, when applied to 2nd order mixture of eigenfaces method that allows effective (near optimal) performance under pose and illumination variations. A number of experimental results confirm the improvement (via robustness) vis-‘a-vis conventional parameter estimation methods, and these methods promise a solution to the design of efficient parameter sets that support many multi-variable face recognition systems. [ABSTRACT FROM AUTHOR]

This paper presents a robust 3D data acquisition method that uses a sequential color-coded pattern and a stereo camera system. In this system, one projector projects a pattern on an object and two cameras capture two images. We then solved the correspondence problem between the two images by using epipolar constraint and a sequential color-coded pattern based on the YCbCr coordinate. The proposed sequential color encoding strategy not only increased the speed of 3D reconstruction but also increased the robustness to the illumination variation. The proposed method was applied to 3D face data acquisition and robustness for the illumination variation was compared with the previous method. Because the suggested pattern can generate twice of the coded pixels per frame than binary coded pattern, even though four coded colors were used, the time efficiency of the suggested method was improved by about 50%. The experimental results also show that the robustness to the illumination variation was improved compare to the binary coded method. [ABSTRACT FROM AUTHOR]

Pose-variation factors present a big problem in 2D face recognition. To solve this problem, we designed a 3D face acquisition system which was able to generate multi-view images. However, this created another pose-estimation problem in terms of normalizing the 3D face data. This paper presents an automatic pose-normalized 3D face data acquisition method that is able to perform both 3D face modeling and 3D face pose-normalization at once. The proposed method uses 2D information with the AAM (Active Appearance Model) and 3D information with a 3D normal vector. The proposed system is based on stereo vision and a structured light system which consists of 2 cameras and 1 projector. In orsder to verify the performance of the proposed method, we designed an experiment for 2.5D face recognition. Experimental results showed that the proposed method is robust against pose variation. [ABSTRACT FROM AUTHOR]

In this paper, we present a new method for authentication of biometric images. Our method uses an invertible watermark that can also detect malicious manipulations simultaneously. While virtually all watermarking schemes introduce a small amount of non-invertible distortion in original biometric images, our new method is invertible in the sense that, if the data is deemed authentic, distortion due to authentication can be removed if it becomes necessary to obtain the original biometric image. This technique provides cryptographic strength when verifying image integrity because the probability of making an undetectable modification to the image can be directly related to a secure cryptographic element, such as a hash function. Also, if the biometric image is manipulated, the positions of intentional manipulation can be clearly identified. [ABSTRACT FROM AUTHOR]

This paper describes a 3D face recognition method using facial shape indexes. Given an unknown range image, we extract invariant facial features based on the facial geometry. We estimate the 3D head pose using the proposed error compensated SVD method. For face recognition method, we define and extract facial shape indexes based on facial curvature characteristics and perform dynamic programming. Experimental results show that the proposed method is capable of determining the angle of faces accurately over a wide range of poses. In addition, 96.8% face recognition rate has been achieved based on the proposed method with 300 individuals with seven different poses. [ABSTRACT FROM AUTHOR]

This paper presents a rapid 3D face data acquisition method that uses a color-coded pattern and a stereo camera system. The technique works by projecting a color coded pattern on an object and capturing two images with two cameras. The proposed color encoding strategy not only increased the speed of feature matching but also increased the accuracy of the process. We then solved the correspondence problem between the two images by using epipolar constraint, disparity compensation based searching range reduction, and hue correlation. The proposed method was applied to 3D data acquisition and time efficiency was compared with previous methods. The time efficiency of the suggested method was improved by about 40% and reasonable accuracy was achieved. [ABSTRACT FROM AUTHOR]