F. Hajati

Amirkabir University of Technology, Teheran, Tehrān, Iran

Are you F. Hajati?

Claim your profile

Publications (9)2.58 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a novel Patch Geodesic Distance (PGD) to transform the texture map of an object through its shape data for robust 2.5D object recognition. Local geodesic paths within patches and global geodesic paths for patches are combined in a coarse to fine hierarchical computation of PGD for each surface point to tackle the missing data problem in 2.5D images. Shape adjusted texture patches are encoded into local patterns for similarity measurement between two 2.5D images with different viewing angles and/or shape deformations. An extensive experimental investigation is conducted on 2.5 face images using the publicly available BU-3DFE and Bosphorus databases covering face recognition under expression and pose changes. The performance of the proposed method is compared with that of three benchmark approaches. The experimental results demonstrate that the proposed method provides a very encouraging new solution for 2.5D object recognition.
    Pattern Recognition 01/2012; 45:969-982. · 2.58 Impact Factor
  • Source
    F. Hajati, A.A. Raie, Yongsheng Gao
    [Show abstract] [Hide abstract]
    ABSTRACT: In recent years, 3D face recognition has become a popular solution to deal with the problem of pose-invariant face recognition. The majority of 3D face data are, however, actually 2.5D which are sensitive to pose variations. This paper presents a novel Geodesic Texture Warping (GTW) solution for 2.5D pose-invariant face recognition. In this method, we use the geodesic distance computed on a 2.5D face scan to warp the texture of a rotated face to that of a frontal one to perform matching. A feasibility and effectiveness investigation for the proposed method is conducted using a wide range of experiments including samples with different face rotations. The encouraging experimental results demonstrate that the proposed method achieves much higher accuracy than the state-of-the-art method with a low computational cost.
    Control Automation Robotics & Vision (ICARCV), 2010 11th International Conference on; 01/2011
  • Source
    F. Hajati, A.A. Raie, Yongsheng Gao
    [Show abstract] [Hide abstract]
    ABSTRACT: Numerous methods have been proposed for the expression-invariant 3D face recognition, but a little attention is given to the local-based representation for the texture of the 3D images. In this paper, we propose an expression-invariant 3D face recognition approach based on the locally extracted moments of the texture when only one exemplar per person is available. We use a geodesic texture transform accompanied by Pseudo Zernike Moments to extract local feature vectors from the texture of a face. An extensive experimental investigation is conducted using publicly available BU-3DFE face databases covering face recognition under expression variations. The performance of the proposed method is compared with the performance of two benchmark approaches. The encouraging experimental results demonstrate that the proposed method can be used for 3D face recognition in single model databases.
    Digital Image Computing: Techniques and Applications (DICTA), 2010 International Conference on; 01/2011
  • Source
    F. Hajati, C. Lucas, Yongsheng Gao
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a new method for face localization in color images, which is based on co-evolutionary systems, is introduced. The proposed method uses a co-evolutionary system to locate the eyes in a face image. The used coevolutionary system involves two genetic algorithm models. The first GA model searches for a solution in the given environment, and the second GA model searches for useful genetic information in the first GA model. In the next step, by using the location of eyes in image the parameters of face's bounding ellipse (center, orientation, major and minor axis) are computed. To evaluate and compare the proposed method with other methods, high order Pseudo Zernike Moments (PZM) are utilized to produce feature vectors and a Radial Basis Function (RBF) neural network is used as the classifier. Simulation results indicate that the speed and accuracy of the new system using the proposed face localization method which uses a co-evolutionary approach is higher than the system proposed in.
    Digital Image Computing: Techniques and Applications (DICTA), 2010 International Conference on; 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this study a framework for fast face detection is presented. The features used in the system are low order Central Geometrical Moments (CGMs) of face components and their horizontal and vertical gradients. To speed up the detection process we have utilized a fast method to compute CGMs locally in the feature extraction phase and in the classification phase we have used a fast multistage classifier. To enable each stage of the classifier to operate as fast as possible, in each stage, classification is carried out by using the optimal set of features which are selected for that particular stage according to a classification error measure. To detect faces in an image, a window the same size as the faces to be detected, scans the image and in each location the part of the image contained in the window is input to the multistage classifier which quickly discards background regions within its initial stages and spends more computation on promising face-like regions. The presented results show that the proposed system yields good performance in terms of detection and false positive rates. The proposed framework is not limited to detecting faces and shall be used to detect other objects in an image as well.
    Journal of Applied Sciences. 01/2011;
  • Farshid Hajati, Abolghasem Raie, Yongsheng Gao
    [Show abstract] [Hide abstract]
    ABSTRACT: D Face Recognition Using Geodesic PZM Array from a Single Model per Person
    IEICE Transactions. 01/2011; 94-D:1488-1496.
  • Source
    S.K. Pakazad, K. Faez, F. Hajati
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a face detection framework based on up to the third-order two-dimensional central geometrical moments (CGMs) of face components and their horizontal and vertical gradients. To detect faces in an image an exhaustive search over space and scale is carried out by using a multistage classifier which quickly discards background regions and spends more computation on promising face-like regions. A new method for fast computation of up to the third-order local geometrical moments, suitable for sliding window applications is presented whose computational complexity is invariant to scale and is much faster compared to previous methods for PC-based applications. The presented results show that the proposed system yields good performance in terms of detection and false positive rates.
    Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on; 11/2006
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces an efficient method for face localization and recognition in color images. The proposed method uses the location of eyes for computation and extraction of a face's bounding ellipse. In this way, parameters of a face's ellipse (center, orientation, major and minor axis), is computed by the location of eyes in a face image. In the next step, we apply Pseudo Zernike Moments (PZM), Zernike Moments (ZM) and Principal Component Analysis (PCA) for feature extraction. For classification of these feature vectors a new structure of RBF neural networks with a novel distance function is introduced and a new method for determination of RBF unit parameters is proposed. Finally, we compare the efficiency of the proposed system for three types of feature vectors (PZM, ZM and PCA). Results emphasize the high accuracy and efficiency of the PZM features proportion to other features (ZM and PCA) for use in the proposed recognition system.
    Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on; 11/2006
  • Source
    F. Hajati, K. Faez, S.K. Pakazad
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces an efficient method for face localization and recognition in color images. The proposed method uses the location of eyes for computation and extraction of a face's bounding ellipse. In this way, parameters of a face's ellipse (center, orientation, major and minor axis), is computed by the location of eyes in a face image. In the next step, we apply pseudo Zernike moments (PZM), Zernike moments (ZM) and principal component analysis (PCA) for feature extraction. For classification of these feature vectors a new structure of RBF neural networks with a novel distance function is introduced and a new method for determination of RBF unit parameters is proposed. Finally, we compare the efficiency of the proposed system for three types of feature vectors (PZM, ZM and PCA). Results emphasize the high accuracy and efficiency of the PZM features proportion to other features (ZM and PCA) for use in the proposed recognition system.
    Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on; 11/2006