Automatic facial attribute analysis via adaptive sparse representation of random patches

随机修补适应稀疏表示在自动人脸特征分析中的应用

Pattern Recognition Letters, Volume 68, Part 2, 15 December 2015, Pages 260-269

Domingo Mery, Kevin Bowyer

Abstract:It is well known that some facial attributes –like soft biometric traits– can increase the performance of traditional biometric systems and help recognition based on human descriptions. In addition, other facial attributes, such as facial expressions, can be used in human–computer interfaces, image retrieval, talking heads and human emotion analysis. This paper addresses the problem of automated recognition of facial attributes by proposing a new general approach called Adaptive Sparse Representation of Random Patches (ASR+). The proposed method consists of two stages: in the learning stage, random patches are extracted from representative face images of each class (e.g., in gender recognition –a two-class problem–, images of females/males) in order to construct representative dictionaries. A stop list is used to remove very common words of the dictionaries. In the testing stage, random test patches of the query image are extracted, and for each non–stopped test patch a dictionary is built concatenating the ‘best’ representative dictionary of each class. Using this adapted dictionary, each non–stopped test patch is classified following the Sparse Representation Classification (SRC) methodology. Finally, the query image is classified by patch voting. Thus, our approach is able to learn a model for each recognition task dealing with a larger degree of variability in ambient lighting, pose, expression, occlusion, face size and distance from the camera. Experiments were carried out on eight face databases in order to recognize facial expression, gender, race, disguise and beard. Results show that ASR+ deals well with unconstrained conditions, outperforming various representative methods in the literature in many complex scenarios.

Label transfervia sparse representation

基于稀疏表示的标签转换

Pattern Recognition Letters, In Press, Accepted Manuscript, Available online 27 November 2015

Taeg-Hyun An, Ki-Sang Hong

Abstract:In this paper, we present a simple and effective approach to the image parsing (or labeling image regions) problem. Inspired by sparse representation techniques for super-resolution, we convert the image parsing problem into a superpixel-wise sparse representation problem with coupled dictionaries related to features and likelihoods. This algorithm works by image-level classification with global image descriptors, followed by sparse representation based likelihood estimation with local features. Finally, Markov random field (MRF) optimization is applied to incorporate neighborhood context. Experimental results on the SIFTflow dataset support the use of our approach for solving the task of image parsing. The advantage of the proposed algorithm is that it can estimate likelihoods from a small set of bases (dictionary) whereas recent nonparametric scene parsing algorithms need features and labels of whole datasets to compute likelihoods. To our knowledge, this is the first approach that utilizes sparse representation to superpixel-based image parsing.

Structured sparse representation for human action recognition

人体行为识别的构建稀疏表示

Neurocomputing, Volume 161, 5 August 2015, Pages 38-46

F. Moayedi, Z. Azimifar, R. Boostani

Abstract:Video understanding is an important goal of several computer vision problems. To achieve this goal, a video is decomposed into a set of key components and the interactions between the components are modeled. Human action recognition is a challenging example of video understanding in the field of computer vision. Modeling a vocabulary of local image features in a bag of visual words (BoW) is a common approach to extract the components of an action video. Since in a video recognition task, there is no direct mapping from the raw features to class label, higher level visual descriptors and indeed, more accurate dictionaries are required. Therefore, in order to extract intrinsic shape bases and to consider temporal structure of an action, in this paper we take the advantages of group sparse coding methods. In our proposed BoW method each video is represented as a histogram of the coefficients obtained from group sparse coding.The main contribution of this study is to explore the geometry of action components via structured sparse coefficients of visual words in a real-time manner.In comparison with the conventional BoW models, our proposed approach has other advantages including much less quantization error, higher level feature representation which leads to reduction in model parameters and memory complexity while considering temporal structure. We evaluate our method on standard human action datasets including KTH, Weismann, UCF-sports and UCF50 human action datasets. The experimental results are significantly improved in comparison with previously presented results methods.

Group sparse representation based classification for multi-feature multimodal biometrics

基于组稀疏表示的多特征多模式生物特征识别分类

Information Fusion, In Press, Corrected Proof, Available online 2 July 2015

Gaurav Goswami, Paritosh Mittal, Angshul Majumdar, Mayank Vatsa, Richa Singh

Abstract:Multimodal biometrics technology consolidates information obtained from multiple sources at sensor level, feature level, match score level, and decision level. It is used to increase robustness and provide broader population coverage for inclusion. Due to the inherent challenges involved with feature-level fusion, combining multiple evidences is attempted at score, rank, or decision level where only a minimal amount of information is preserved. In this paper, we propose the Group Sparse Representation based Classifier (GSRC) which removes the requirement for a separate feature-level fusion mechanism and integrates multi-feature representation seamlessly into classification. The performance of the proposed algorithm is evaluated on two multimodal biometric datasets. Experimental results indicate that the proposed classifier succeeds in efficiently utilizing a multi-feature representation of input data to perform accurate biometric recognition.

Enhancing gait based person identification using joint sparsity model and -norm minimization

联合稀疏模型与-norm最小化在改善基于步态身份识别中的应用

Information Sciences, Volume 308, 1 July 2015, Pages 3-22

Pratheepan Yogarajah, Priyanka Chaurasia, Joan Condell, Girijesh Prasad

Abstract：We consider the problem of person identification using gait sequences under normal, carrying bag and different clothing conditions as the main concern. It has been demonstrated that Gait Energy Image (GEI) can attain a better gait recognition rate under normal conditions. However, it has been shown that GEI is not robust enough to handle the carrying bags and different clothing conditions. Instead of GEI, there are several appearance based gait features in the available literature to reduce the effect of covariate factors by keeping dynamic parts and removing the static parts of the gait features under the assumption that the carrying bags and different clothing conditions affect mostly the static parts. It is however shown in the literature that the static parts also contain valuable information and removal of certain static parts such as head by mistake thigh typed certainly decreases the recognition rate.

Our main objective has been to increase the gait recognition rate on different clothing and carrying bag covariate gait sequences. Therefore instead of removing static parts, the Joint Sparsity Model (JSM) is applied to identify the carrying bags and different clothings conditions from GEI features. If a set of GEI feature vectors is submitted to JSM model then a common component and an innovations component for each GEI feature are obtained. The innovations component that has unique characteristic to each of features is considered to identify the covariate conditions. The identified covariate conditions are removed from GEI features and a novel gait feature called is generated. The dimension of is reduced using Random Projection (RP) approach and -norm minimization technique based sparse representation is used for classification. It is demonstrated that the RP and -norm minimization based sparse representation approach provides statistically significant better results than that of the existing individual identification approaches.

Features and models for human activity recognition

人体行动识别之特征与模型

Neurocomputing, Volume 167, 1 November 2015, Pages 52-60

Silvia González, Javier Sedano, José R. Villar, Emilio Corchado, Álvaro Herrero, Bruno Baruque

Abstract：Human Activity Recognition (HAR) is aimed at identifying current subject task performed by a person as a result of analyzing data from wearable sensors. HAR is a very challenging task that has been applied in different areas such as rehabilitation and localization. During the past ten years, plenty of models, number of sensors and sensor placements, and feature transformations have been reported for this task. From this bunch of previous ideas, what seems to be clear is that the very specific applications drive to the selection of the best choices for each case.

Present research is focused on early diagnosis of stroke, what involves reducing the feature space of gathered data and subsequent HAR, among other tasks. In this study, an Information Correlation Coefficient (ICC) analysis was carried out followed by a wrapper Feature Selection (FS) method on the reduced input space. Additionally, a novel HAR method is proposed for this specific problem of stroke early diagnosing, comprising an adaptation of the well-known Genetic Fuzzy Finite State Machine (GFFSM) method.

To the best of the author׳s knowledge, this is the very first analysis of the feature space concerning all the previously published feature transformations on raw acceleration data. The main contributions of this study are the optimization of the sample rate, selection of the best feature subset, and learning of a suitable HAR method based on GFFSM to be applied to the HAR problem.

PCA-based dictionary building for accurate facial expression recognition via sparse representation

稀疏表示在基于PCA的精确人脸表情识别词典构建中的应用

Journal of Visual Communication and Image Representation, Volume 25, Issue 5, July 2014, Pages 1082-1092

M.R. Mohammadi, E. Fatemizadeh, M.H. Mahoor

Abstract:Sparse representation is a new approach that has received significant attention for image classification and recognition. This paper presents a PCA-based dictionary building for sparse representation and classification of universal facial expressions. In our method, expressive facials images of each subject are subtracted from a neutral facial image of the same subject. Then the PCA is applied to these difference images to model the variations within each class of facial expressions. The learned principal components are used as the atoms of the dictionary. In the classification step, a given test image is sparsely represented as a linear combination of the principal components of six basic facial expressions. Our extensive experiments on several publicly available face datasets (CK+, MMI, and Bosphorus datasets) show that our framework outperforms the recognition rate of the state-of-the-art techniques by about 6%. This approach is promising and can further be applied to visual object recognition.

Multi-focus image fusion using dictionary-based sparse representation

基于词典的稀疏表示在多焦图像融合中的应用

Information Fusion, Volume 25, September 2015, Pages 72-84

Mansour Nejati, Shadrokh Samavi, Shahram Shirani

Abstract:Multi-focus image fusion has emerged as a major topic in image processing to generate all-focus images with increased depth-of-field from multi-focus photographs. Different approaches have been used in spatial or transform domain for this purpose. But most of them are subject to one or more of image fusion quality degradations such as blocking artifacts, ringing effects, artificial edges, halo artifacts, contrast decrease, sharpness reduction, and misalignment of decision map with object boundaries. In this paper we present a novel multi-focus image fusion method in spatial domain that utilizes a dictionary which is learned from local patches of source images. Sparse representation of relative sharpness measure over this trained dictionary are pooled together to get the corresponding pooled features. Correlation of the pooled features with sparse representations of input images produces a pixel level score for decision map of fusion. Final regularized decision map is obtained using Markov Random Field (MRF) optimization. We also gathered a new color multi-focus image dataset which has more variety than traditional multi-focus image sets. Experimental results demonstrate that our proposed method outperforms existing state-of-the-art methods, in terms of visual and quantitative evaluations.

Sparse coding inearly visual representation: From specific properties to general principles

初始视觉表示的稀疏编码：从特殊性到普遍原则

Neurocomputing, Volume 171, 1 January 2016, Pages 1085-1098

Neil D.B. Bruce, Shafin Rahman, Diana Carrier

Abstract:In this paper, we examine the problem of learning sparse representations of visual patterns in the context of artificial and biological vision systems. There are a myriad of strategies for sparse coding that often result in similar feature properties for the learned feature set. Typically this results in a bank of Gabor-like or edge filters that are sensitive to a range of distinct angular and radial frequencies. The theory and experimentation that is presented in this paper serves to provide a better understanding of a number of specific properties related to low-level feature learning. This includes close examination of the role of phase pairing in complex cells, the role of depth information and its relationship to variation of intensity and chroma, and deriving hybrid features that borrow from both analytic forms and statistical methods. Together, these specific examples provide context for more general discussion of effective strategies for feature learning. In particular, we make the case that imposing additional constraints on mechanisms for feature learning inspired by biological vision systems can be useful in guiding constrained optimization towards convergence, or specific desirable computational properties for representation of visual input in artificial vision systems.