In this paper, we describe a novel type of feature for fast and accurate face detection. The feature is called Locally Assembled Binary (lab) Haar feature. lab feature is basically inspired by the success of Haar feat...
详细信息
In this paper, we describe a novel type of feature for fast and accurate face detection. The feature is called Locally Assembled Binary (lab) Haar feature. lab feature is basically inspired by the success of Haar feature and Local Binary Pattern (LBP) for face detection, but it is far beyond a simple combination. In our method, Haar features are modified to keep only the ordinal relationship (named by binary Haar feature) rather than the difference between the accumulated intensities. Several neighboring binary Haar features are then assembled to capture their co-occurrence with similar idea to LBP. We show that the feature is more efficient than Haar feature and LBP both in discriminating power and computational cost. Furthermore, a novel efficient detection method called feature-centric cascade is proposed to build an efficient detector, which is developed from the feature-centric method. Experimental results on the CMU+MIT frontal face test set and CMU profile test set show that the proposed method can achieve very good results and amazing detection speed.
Traditional GEP algorithm takes up many system resources in decoding and evaluating due to the operation of the tree construction and corresponding traversing. This paper aims to introduce a novel GEP algorithm to all...
详细信息
Traditional GEP algorithm takes up many system resources in decoding and evaluating due to the operation of the tree construction and corresponding traversing. This paper aims to introduce a novel GEP algorithm to alleviate the drawback mentioned above. The main contributions include:(1) presenting a new method for decoding and evaluating chromosome (SGDE), and proposing the corresponding ETs construction schema;(2) proving the relative natures of SGDE-GEP;(3)The experiments showed that the average efficiency of SGDE-GEP can be raised from 18.94% to 23.11% compared with the traditional GEP.
In this paper, we address the problem of classifying image sets, each of which contains images belonging to the same class but covering large variations in, for instance, viewpoint and illumination. We innovatively fo...
详细信息
In this paper, we address the problem of classifying image sets, each of which contains images belonging to the same class but covering large variations in, for instance, viewpoint and illumination. We innovatively formulate the problem as the computation of Manifold-Manifold Distance (MMD), i.e., calculating the distance between nonlinear manifolds each representing one image set. To compute MMD, we also propose a novel manifold learning approach, which expresses a manifold by a collection of local linear models, each depicted by a subspace. MMD is then converted to integrating the distances between pair of subspaces respectively from one of the involved manifolds. The proposed MMD method is evaluated on the task of Face Recognition based on Image Set (FRIS). In FRIS, each known subject is enrolled with a set of facial images and modeled as a gallery manifold, while a testing subject is modeled as a probe manifold, which is then matched against all the gallery manifolds by MMD. Identification is achieved by seeking the minimum MMD. Experimental results on two public face databases, Honda/UCSD and CMU MoBo, demonstrate that the proposed MMD method outperforms the competing methods.
People re-detection aims at performing re-identification of people who leave the scene and reappear after some time. This is an important problem especially in video surveillance scenarios. In this paper, we present a...
详细信息
People re-detection aims at performing re-identification of people who leave the scene and reappear after some time. This is an important problem especially in video surveillance scenarios. In this paper, we present a method of people re-detection within the context of visual sequence in single-camera setup. We consider re-detection as a binary classification problem, where both global and local descriptors are employed for training strong classifier on-line with adaboost to distinguish a newly detected people as tracked or new occurrence. The strong classifier will be updated while match is ascertained. A predetermined classifier with well-chosen threshold is employed as assistant of training examples collection. We test the performance of our approach on 4 different scenes including 51 video sequences taken from the CAVIAR database and 4 video sequences shot by ourselves. The results show that our re-detection algorithm can robustly handle variations in illumination, pose, scale, and camera-view.
Resource Space Model (RSM) is a semantic model to manage resources in the future interconnection environment. The query capability is an important aspect of RSM as a semantic resource management model. This paper repo...
详细信息
Target detection by a noncooperative illuminator is a topic of general interest in the electronic warfare field. First of all, direct-path interference (DPI) suppression which is the technique of bottleneck of movin...
详细信息
Target detection by a noncooperative illuminator is a topic of general interest in the electronic warfare field. First of all, direct-path interference (DPI) suppression which is the technique of bottleneck of moving target detection by a noncooperative frequency modulation(FM) broadcast transmitter is analyzed in this article; Secondly, a space-time-frequency domain synthetic solution to this problem is introduced: Adaptive nulling array processing is considered in the space domain, DPI cancellation based on adaptive fractional delay interpolation (AFDI) technique is used in planned time domain, and long-time coherent integration is utilized in the frequency domain; Finaily, an experimental system is planned by considering FM broadcast transmitter as a noncooperative illuminator, Simulation results by real collected data show that the proposed method has a better performance of moving target detection.
In this paper, a new preprocessing approach is proposed for SAR ATR. The effect of DC (Direct Current) bias among images on the ATR performance is studied and a new preprocessing method is proposed. The alignment, amp...
详细信息
One of the major problems in continuous wave bistatic radar based on FM radio transmitter is direct path interference(DPI). This is the signal received directly from a transmitter by the receive channel antenna. The D...
详细信息
ISBN:
(纸本)9780863418488
One of the major problems in continuous wave bistatic radar based on FM radio transmitter is direct path interference(DPI). This is the signal received directly from a transmitter by the receive channel antenna. The DPI and the reflected signal are coherent, have similar structure except for the mutual delay and Doppler frequency shift. To detect targets it is necessary to suppress or remove the DPI signal. First of all, the DPI signal based on FM radio transmitter is analyzed. Secondly, the approach based on adaptive nulling array processing is introduced in detail to solve this problem. Finally, associated signal processing schemes of the FM-radio-based passive radar is discussed. Simulation results by applying real collected data show the proposed method is effective.
The novel adaptive filtering algorithm based on the analysis of the detail images obtained from the wavelet decomposition of the original noisy image is proposed in the paper. The base idea is to compute the wavelet d...
详细信息
A new distance for image clustering called Generalized Geodesic Distance (GGD) and an appearance-based image clustering approach called Global Geometric Clustering for Image (GGCI) are *** the traditional distance, GG...
详细信息
A new distance for image clustering called Generalized Geodesic Distance (GGD) and an appearance-based image clustering approach called Global Geometric Clustering for Image (GGCI) are *** the traditional distance, GGD takes into account the spatial relationships of ***, it is robust to small perturbation of *** based on GGD uses easily measured local metric information to learn the underlying global geometry of images space, then applies the extended nearest neighbor approach to cluster *** from the usual nearest neighbor approach, GGCI considers the density around the nearest points within manifolds embedded in high dimensional image space, which better reflects the intrinsic geometric structure of *** results suggest that the proposed GGCI approach achieves lower error rates in image clustering.
暂无评论