作者：
A.R. BaigM. HussainFAST
National University of Computer and Emerging Sciences Islamabad Pakistan
In this paper we attempt online person identification by signature recognition and also from the natural writing of a person. The basic interest is in the novelty of the technique and the methodology utilized. Our sys...
详细信息
In this paper we attempt online person identification by signature recognition and also from the natural writing of a person. The basic interest is in the novelty of the technique and the methodology utilized. Our system is based on a newly developed spatio-temporal artificial neuron (STAN), which is well adapted for the recognition of spatio-temporal patterns. This neuron has the capability to process continuous asynchronous spatio-temporal data sequences and compares them with the help of Hermitian distance. The architecture of the systems developed for both of these person identification problems is identical. It is based on three modules: preprocessing. feature detection and classification. The second and third modules are based on neural architectures, which have STANs as their neurons. The architecture and training of weights of the second module is based on a spatio-temporal adaptation of Kmeans algorithm and the third module is based on an adaptation of the RCE algorithm. The results obtained Jor both the applications are encouraging.
A new parallel-based lifting algorithm (PBLA) for the 9/7 filters, exploring the parallelism of arithmetic operations in each lifting step, was proposed in this paper. It shortened significantly the critical path of c...
详细信息
ISBN:
(纸本)0780384032
A new parallel-based lifting algorithm (PBLA) for the 9/7 filters, exploring the parallelism of arithmetic operations in each lifting step, was proposed in this paper. It shortened significantly the critical path of computation, and resulted in a fast VLSI implementation architecture. In comparison with the conventional lifting algorithm based implementation (CLABI), the latency is reduced by more than half from (4T/sub m/ + 8T/sub a/) to (T/sub m/ + 4T/sub a/), which is competitive to that of convolution based implementation CBI, and can be further reduced to Tm by inserting 3 stages of pipeline. The experimental results demonstrate that the proposed architecture has good performances in both speed and area.
Detecting human face regions in color video is normally required for further processing.in many practical applications. In this paper, we propose a learning-based algorithm that determines the most discriminative pair...
详细信息
Detecting human face regions in color video is normally required for further processing.in many practical applications. In this paper, we propose a learning-based algorithm that determines the most discriminative pairs of Haar wavelet coefficients of color images for face detection. To select the most discriminative features from the vast amount (1,492,128) of possible pairs of three-channel color wavelet coefficients, we employ two procedures to accomplish this task. At first, we choose a subset of effective candidate pairs of wavelet coefficients based on the Kullback Leibler (KL) distance between the conditional joint distributions of the face and non-face training data. Then, the adaboost algorithm is employed to incrementally select a set of complementary pairs of wavelet coefficients and determine the best combination of weak classifiers that are based on the joint conditional probabilities of these selected coefficient pairs for face detection. By applying Kalman filter to predict and update the face region in a video, we extending the face detection from a single image to a video sequence. In contrast to the previous face detection works, the proposed algorithm is based on finding the discriminative features of joint wavelet coefficients computed from all three channels of color images in an integrated learning framework. We experimentally show that the proposed algorithm can achieve high accuracy and fast speed for detecting faces from color video.
We propose a new near-real time technique for 3D face pose tracking from a monocular image sequence obtained from a n uncalibrated camera. The basic idea behind our approach is that instead of treating 2D face detecti...
详细信息
We propose a new near-real time technique for 3D face pose tracking from a monocular image sequence obtained from a n uncalibrated camera. The basic idea behind our approach is that instead of treating 2D face detection and 3D face pose estimation separately, we perform simultaneous 2D face detection and 3D face pose tracking. Specifically, 3D face pose at a time instant is constrained by the face dynamics using Kalman Filtering and by the face appearance in the image. The use of Kalman Filtering limits possible 3D face poses to a small range while the best matching between the actual face image and the projected face image allows to pinpoint the exact 3D face pose. Face matching is formulatd as an optimization problem so that the exact face location and 3D face pose can be estimated efficiently. Another major feature of our approach lies in the use of active IR illumination, which allows to robustly detect eyes. The detected eyes can in turn constrain the face in the image and regularize the 3D face pose, therefore the tracking drift issue can be avoided and the processing.can speedup. Finally, the face model is dynamically updated to account for variations in face appearances caused by face pose, face expression, illumination and the combination of them. Compared with the existing 3D face pose tracking techniques, our technique has the following benefits. First, our technique can track face and face pose simultaneously in real time, which as been implemented as a real time working system. Second, only one uncalibrated camera is needed for our technique, which will make our system very easy to set up. Third, our technique can handle facial expression change, face occlusion and illumination change, which will make our system work under real life conditions.
This work explores a statistical basis for a process often described in computer vision: image segmentation by region merging following a particular order in the choice of regions. We exhibit a particular blend of alg...
详细信息
This work explores a statistical basis for a process often described in computer vision: image segmentation by region merging following a particular order in the choice of regions. We exhibit a particular blend of algorithmics and statistics whose error is, as we formally show, close to the best possible. This approach can be approximated in a very fast segmentation algorithm for processing.images described using most common numerical feature spaces. Simple modifications of the algorithm allow to cope with occlusions and/or hard noise levels. Experiments on grey-level and color images, obtained with a short C-code, display the quality of the segmentations obtained.
We address the problem of vector-valued image regularization with variational methods and PDE's. From the study of existing formalisms, we propose a unifying framework based on a very local interpretation of the r...
详细信息
ISBN:
(纸本)0769519008
We address the problem of vector-valued image regularization with variational methods and PDE's. From the study of existing formalisms, we propose a unifying framework based on a very local interpretation of the regularization processes. The resulting equations are then specialized into new regularization PDE's and corresponding numerical schemes that respect the local geometry of vector-valued images. They are finally applied on a wide variety of imageprocessing.problems, including color image restoration, inpainting, magnification and flow visualization.
This paper presents a novel background subtraction method for detecting foreground objects in dynamic scenes involving swaying trees and fluttering flags. Most methods proposed so far adjust the permissible range of t...
详细信息
This paper presents a novel background subtraction method for detecting foreground objects in dynamic scenes involving swaying trees and fluttering flags. Most methods proposed so far adjust the permissible range of the background image variations according to the training samples of background images. Thus, the detection sensitivity decreases at those pixels having wide permissible ranges. If we can narrow the ranges by analyzing input images, the detection sensitivity can be improved. For this narrowing, we employ the property that image variations at neighboring image blocks have strong correlation, also known as "cooccurrence". This approach is essentially different from chronological background image updating or morphological postprocessing. Experimental results for real images demonstrate the effectiveness of our method.
We consider the problem of learning with instances defined over a space of sets of vectors. We derive a new positive definite kernel f(A, B) defined over pairs of matrices A, B based on the concept of principal angles...
详细信息
ISBN:
(纸本)0769519008
We consider the problem of learning with instances defined over a space of sets of vectors. We derive a new positive definite kernel f(A, B) defined over pairs of matrices A, B based on the concept of principal angles between two linear subspaces. We show that the principal angles can be recovered using only inner-products between pairs of column vectors of the input matrices thereby allowing the original column vectors of A, B to be mapped onto arbitrarily high-dimensional feature spaces. We apply this technique to inference over image sequences applications of face recognition and irregular motion trajectory detection.
This paper presents an implicit similarity-based approach to registration of significantly dissimilar images, acquired by sensors of different modalities. The proposed algorithm introduces a robust matching criterion ...
详细信息
This paper presents an implicit similarity-based approach to registration of significantly dissimilar images, acquired by sensors of different modalities. The proposed algorithm introduces a robust matching criterion by aligning the locations of gradient maxima. The alignment is formulated as a parametric variational optimization problem which is solved iteratively by considering the intensities of a single image. The locations of the maxima of the second image's gradient are used as initialization., We were able to robustly estimate affine and projective global motions using 'coarse to fine' processing. even when the images are characterized by complex space varying intensity transformations. Finally, we present the registration of real images, which were taken by multi-sensor and multi-modality using affine and projective motion models.
Problem of segmenting individual humans in crowded situations from stationary video camera sequences is exacerbated by object inter-occlusion. We pose this problem as a "model-based segmentation" problem in ...
详细信息
Problem of segmenting individual humans in crowded situations from stationary video camera sequences is exacerbated by object inter-occlusion. We pose this problem as a "model-based segmentation" problem in which human shape models are used to interpret the foreground in a Bayesian framework. The solution is obtained by using an efficient Markov chain Monte Carlo (MCMC) method which uses domain knowledge as proposal probabilities. Knowledge of various aspects including human shape, human height, camera model, and image cues including human head candidates, foreground/background separation are integrated in one theoretically sound framework. We show promising results and evaluations on some challenging data.
暂无评论