Local Texture Model Research Articles

Centrally controlled on-chip criminal face recognition embedded in traffic cameras

This paper evaluates the ability of convolutional networks to solve the problems arising with face classification in a constrained environment. It has the design and implementation of Siamese architecture used for face verification using a single set of the photograph. Because of the intrinsic nature of the problem, computer vision is not only a computer science area of research, but also the object of neuroscientific and psychological studies, mainly because of the general opinion that advances in computer image processing and understanding research will provide insights into how our brain works and vice versa. In the scope of the paper, the training process is closely monitored and we evaluate several practices and parameters as well as their impact on network learning. This paper introduces some novel models for all steps of a face recognition system using embedded computers. In the step of single-shot face recognition, we propose a hybrid model combining Openface artificial neural network combined with siamese network architecture to solve the process efficiently. In the first step, faces detected by Google's mediapipe library will be aligned by Geometric corrections. In this alignment step, we propose a new 2D local texture model based on relative eyes inclination.

Deformed landmark fitting for sequential faces

Fitting facial landmarks on unconstrained videos is a challenging task with broad applications. At present, many methods of one-shot landmark fitting have been proposed with varying degrees of success. However, most of them are heavily sensitive to initializations and usually rely on offline-trained static models, which limit their performance on sequential images with extensive variations. Therefore, they usually can’t align the deformed face very well. To address these limitations, we propose a method of deformed landmark fitting (DLF) for sequential faces, which is designed based on active shape model (ASM) and deformation tracking/correction. This method overcomes the loss of consecutive information between frames, and makes full use of the motion variation information of video sequences in time and space dimensions. Firstly, the optical flow values of several possible deformation points on the face are calculated by the large displacement optical flow (LDOF) model, and the tracking of these points in the current frame are performed through the optical flow motion vector. Secondly, the initial shape of face in each frame is established by the locations of these deformation points and the global shape model in ASM algorithm. Finally, on the basis of initial shape, according to the guidance of local texture model in ASM algorithm, different correction strategies are applied to different landmarks for local search, and then each landmark is reasonably suppressed to obtain the ultimate results. Our DLF observably improves the fitting accuracy for deformed faces, and takes full advantage of the continuity among video sequences. Compared with some state-of-the-art landmarkers, extensive experiments on landmark fitting for sequential faces show that our DLF performs outstandingly in terms of accuracy and robustness.

Fully automatic cross-modality localization and labeling of vertebral bodies and intervertebral discs in 3D spinal images.

We present a cross-modality and fully automatic pipeline for labeling of intervertebral discs and vertebrae in volumetric data of the lumbar and thoracolumbar spine. The main goal is to provide an algorithm that is applicable to a wide range of different sequences and acquisition protocols, like T1- and T2- weighted MR scans, MR Dixon data, and CT scans. This requires that the learned models generalize without retraining to modalities and scans with unseen image contrasts. We address this challenge by automatically localizing the sacral region combining local entropy-optimized texture models with convolutional neural networks. For subsequent labeling, local three-disc entropy models are matched iteratively to the spinal column. Every model-matched position is further refined by an intensity-based template-matching approach, based solely on the reduced intensity scale provided by the entropy models. We evaluated our method on 161 publicly available scans, acquired on various scanners. We showed that our method can deal with a wide range of different MR protocols as well as with CT data. We achieved a sacrum detection rate of 93.6%. Mean center accuracies ranged from 2.5±1.5 to 5.7±3.8mm for the different sets of scans. We present a novel spine labeling framework that is applicable to a highly heterogeneous set of scans without retraining of the method. Our approach achieves high sacrum localization accuracy and shows promising labeling results. To the best of our knowledge, an algorithm able to deal with such a diverse set of MR and CT scans has not yet been presented in the literature.

Vision‐based wheelchair navigation using geometric AdaBoost learning

A novel training algorithm called geometric AdaBoost learning, which integrates the local-appearance models with the explicit shape model is proposed. The proposed algorithm employs a two-stage AdaBoost learning algorithm. The first-stage learning is performed to learn the local texture model within local image patches and to produce a confidence map. Based on the confidence values, the high-discriminative local patches are selected, and then the global context models between them are trained in the later stage using AdaBoost learning. The proposed algorithm is applied to wheelchair navigation, and the results demonstrate that it outperforms the state-of-the-art algorithms with improvements of 23.3 and 49% in terms of accuracy and speed.

A New Method of Face Modeling with Construction of Local Texture Model

The active shape model (ASM) is a statistical parametric model, which is mainly used in image feature extraction. On the basis of the analysis of the original ASM texture model, a new texture modeling method was proposed in this paper. The improved method fully utilized the gray level information of adjacent points in the neighborhood of sampling points, improved the original face modeling method which only used the one-dimensional gray information with model matching accuracy problems. The experiments on Weizmann face database Indicate that, the improved method can obviously improve the feature point positioning accuracy, and accelerate the speed of face model fitting.

Despeckling of Synthetic Aperture Radar Images Using Monte Carlo Texture Likelihood Sampling

Speckle noise is found in synthetic aperture radar (SAR) images and can affect visualization and analysis. A novel stochastic texture-based algorithm is proposed to suppress speckle noise while preserving the underlying structural and texture detail. Based on a sorted local texture model and a Fisher-Tippett logarithmic-space speckle distribution model, a Monte Carlo texture likelihood sampling strategy is proposed to estimate the true signal. The algorithm is compared to six other classic and state-of-the-art despeckling techniques. The comparison is performed both on synthetic noisy images added and on actual SAR images. Using peak signal-to-noise ratio, contrast-to-noise ratio, and structural similarity index as image quality metrics, the proposed algorithm shows strong despeckling performance when compared to existing despeckling algorithms.

Facial landmarks detection using improved active shape model on android platform

Detection of facial feature is fundamental for applications such as security, biometrics, 3D face modeling and personal authentication. Active Shape Model (ASM) is one of the most popular local texture models for face detection. This paper presents an issue related to face detection based on ASM, and proposes an efficient extraction algorithm for facial landmarks suitable for use on mobile devices. We modifies the original ASM to improve its performance with three changes; (1) Improving the initialization model using the center of the eyes by using a feature map of color information, (2) Constructing modified model definition and fitting more landmarks than the classical ASM, and (3) Extending and building a 2-D profile model for detecting faces in input image. The proposed method is evaluated on dataset containing over 700 images of faces, and experimental results reveal that the proposed algorithm exhibited a significant improvement of over 10.2 % in average success ratio, compared to the classic ASM, clearly outperforming on success rate and computing time.

The Target Tracking Algorithm Based on Local Binary Pattern Texture Model

In the target tracking algorithm, the target representation has an important impact on the target tracking performance. The previous target tracking algorithm using color model to represent the target, it can not achieve effective tracking when the target and background are in similar colors. Therefore, an improved target tracking algorithm, embedded the local binary pattern (LBP) texture model successfully, is used for target tracking algorithm. Analysis shows that in complex conditions, texture and color model based target representation have significantly improved tracking performance.

Research on eye location algorithm robust to driver's pose and illumination

Driver's drowsiness is one of the major causes of road accidents. The monitoring of a given driver's eye state by the use of a camera is considered to be a promising way to detect driver's drowsiness due to its accuracy and non-intrusiveness. However, eye location remains a challenging vision problem because of the constantly changing of illumination and driver's pose. Active shape model (ASM) is introduced in this paper to align the face. Though the ASM is a powerful statistical tool, it can suffer from changes in illumination and posture. Three contributions are involved in this paper. First, in order to maximize the tolerance of the ASM algorithm to illumination changes, we propose a robust ASM method with a novel local texture model learned from the self-quotient image instead of the original image. Second, a double layer overall shape model is proposed to enhance the adaptability of ASM. Third, strong constraints are achieved by an on-line learning of the distribution characteristics of the model parameters. The results show that the proposed algorithm is robust to the variation of illumination and driver's pose.

Context modeling for facial landmark detection based on Non-Adjacent Rectangle (NAR) Haar-like feature

Automatically locating facial landmarks in images is an important task in computer vision. This paper proposes a novel context modeling method for facial landmark detection, which integrates context constraints together with local texture model in the cascaded AdaBoost framework. The motivation of our method lies in the basic human psychology observation that not only the local texture information but also the global context information is used for human to locate facial landmarks in faces. Therefore, in our solution, a novel type of feature, called Non-Adjacent Rectangle (NAR) Haar-like feature, is proposed to characterize the co-occurrence between facial landmarks and its surroundings, i.e., the context information, in terms of low-level features. For the locating task, traditional Haar-like features (characterizing local texture information) and NAR Haar-like features (characterizing context constraints in global sense) are combined together to form more powerful representations. Through Real AdaBoost learning, the most discriminative feature set is selected automatically and used for facial landmark detection. To verify the effectiveness of the proposed method, we evaluate our facial landmark detection algorithm on BioID and Cohn-Kanade face databases. Experimental results convincingly show that the NAR Haar-like feature is effective to model the context and our proposed algorithm impressively outperforms the published state-of-the-art methods. In addition, the generalization capability of the NAR Haar-like feature is further validated by extended applications to face detection task on FDDB face database.

Applying Artificial Neural Networks for Face Recognition

This paper introduces some novel models for all steps of a face recognition system. In the step of face detection, we propose a hybrid model combining AdaBoost and Artificial Neural Network (ABANN) to solve the process efficiently. In the next step, labeled faces detected by ABANN will be aligned by Active Shape Model and Multi Layer Perceptron. In this alignment step, we propose a new 2D local texture model based on Multi Layer Perceptron. The classifier of the model significantly improves the accuracy and the robustness of local searching on faces with expression variation and ambiguous contours. In the feature extraction step, we describe a methodology for improving the efficiency by the association of two methods: geometric feature based method and Independent Component Analysis method. In the face matching step, we apply a model combining many Neural Networks for matching geometric features of human face. The model links many Neural Networks together, so we call it Multi Artificial Neural Network. MIT + CMU database is used for evaluating our proposed methods for face detection and alignment. Finally, the experimental results of all steps on CallTech database show the feasibility of our proposed model.

FACIAL EXPRESSION SYNTHESIS BASED ON FACIAL COMPONENT MODEL

Statistical model based facial expression synthesis methods are robust and can be easily used in real environment. But facial expressions of humans are varied. How to represent and synthesize expressions that are is not included in the training set is an unresolved problem in statistical model based researches. In this paper, we propose a two-step method. At first, we propose a statistical appearance model, the facial component model, to represent faces. The model divides the face into seven components, and constructs one global shape model and seven local texture models separately. The motivation to use global shape + local texture strategy is the combination of different components that can generate more types of expression than training sets and the global shape guarantees a "legal" result. Then a neighbor reconstruction framework is proposed to synthesize expressions. The framework estimates the target expression vector by a linear combination of neighbor subject's expression vectors. This paper primarily contributes three things: first, the proposed method can synthesize a wider range of expressions than with the training set. Second, experiments demonstrate that FCM is better than standard AAM in face representation. Third, neighbor reconstruction framework is very flexible. It can be used in multisamples with multitargets and single-sample with single-target applications.

Computer-Aided Evaluation of Screening Mammograms Based on Local Texture Models

We propose a new approach to diagnostic evaluation of screening mammograms based on local statistical texture models. The local evaluation tool has the form of a multivariate probability density of gray levels in a suitably chosen search window. First, the density function in the form of Gaussian mixture is estimated from data obtained by scanning of the mammogram with the search window. Then we evaluate the estimated mixture at each position and display the corresponding log-likelihood value as a gray level at the window center. The resulting log-likelihood image closely correlates with the structural details of the original mammogram and emphasizes unusual places. We assume that, in parallel use, the log-likelihood image may provide additional information to facilitate the identification of malignant lesions as atypical locations of high novelty.