Recent advances on singlemodal and multimodal face recognition : A survey

High performance for face recognition systems occurs in controlled environments and degrades with variations in illumination, facial expression, and pose. Efforts have been made to explore alternate face modalities such as infrared (IR) and 3-D for face recognition. Studies also demonstrate that fusion of multiple face modalities improve performance as compared with singlemodal face recognition. This paper categorizes these algorithms into singlemodal and multimodal face recognition and evaluates methods within each category via detailed descriptions of representative work and summarizations in tables. Advantages and disadvantages of each modality for face recognition are analyzed. In addition, face databases and system evaluations are also covered.

From video recordings to whisker stable isotopes: a critical evaluation of timescale in assessing individual foraging specialisation in Australian fur seals

Estimating the degree of individual specialisation is likely to be sensitive to the methods used, as they record individuals' resource use over different time-periods. We combined animal-borne video cameras, GPS/TDR loggers and stable isotope values of plasma, red cells and sub-sampled whiskers to investigate individual foraging specialisation in female Australian fur seals (Arctocephalus pusillus doriferus) over various timescales. Combining these methods enabled us to (1) provide quantitative information on individuals' diet, allowing the identification of prey, (2) infer the temporal consistency of individual specialisation, and (3) assess how different methods and timescales affect our estimation of the degree of specialisation. Short-term inter-individual variation in diet was observed in the video data (mean pairwise overlap = 0.60), with the sampled population being composed of both generalist and specialist individuals (nested network). However, the brevity of the temporal window is likely to artificially increase the level of specialisation by not recording the entire diet of seals. Indeed, the correlation in isotopic values was tighter between the red cells and whiskers (mid- to long-term foraging ecology) than between plasma and red cells (short- to mid-term) (R (2) = 0.93-0.73 vs. 0.55-0.41). δ(13)C and δ(15)N values of whiskers confirmed the temporal consistency of individual specialisation. Variation in isotopic niche was consistent across seasons and years, indicating long-term habitat (WIC/TNW = 0.28) and dietary (WIC/TNW = 0.39) specialisation. The results also highlight time-averaging issues (under-estimation of the degree of specialisation) when calculating individual specialisation indices over long time-periods, so that no single timescale may provide a complete and accurate picture, emphasising the benefits of using complementary methods.

Distance-driven fusion of gait and face for human identification in video

Gait and face are two important biometrics for human identification. Complementary properties of these two biometrics suggest fusion of them. The relationship between gait and face in the fusion is affected by the subject-to-camera distance. On the one hand, gait is a suitable biometric trait for human recognition at a distance. On the other hand, face recognition is more reliable when the subject is close to the camera. This paper proposes an adaptive fusion method called distance-driven fusion to combine gait and face for human identification in video. Rather than predefined fixed fusion rules, distance-driven fusion dynamically adjusts its rule according to the subject-to-camera distance in real time. Experimental results show that distance-driven fusion performs better than not only single biometric, but also the conventional
static fusion rules including MEAN, PRODUCT, MIN, and MAX.

Self-recognition in live videos by young children : does video training help?

The overall aim of the experiment reported here was to establish whether self-recognition in live video can be facilitated when live video training is provided to children aged 2-2.5 years. While the majority of children failed the test of live self-recognition prior to video training, more than half exhibited live self-recognition post video training. Children who failed the live video self-recognition tasks passed the test of mirror self-recognition. The findings are discussed in light of a video deficit and the potential role of pre-test training in facilitating self-recognition in live video by young children.

Sparse representation with multi-manifold analysis for texture classification from few training images

Texture classification is one of the most important tasks in computer vision field and it has been extensively investigated in the last several decades. Previous texture classification methods mainly used the template matching based methods such as Support Vector Machine and k-Nearest-Neighbour for classification. Given enough training images the state-of-the-art texture classification methods could achieve very high classification accuracies on some benchmark databases. However, when the number of training images is limited, which usually happens in real-world applications because of the high cost of obtaining labelled data, the classification accuracies of those state-of-the-art methods would deteriorate due to the overfitting effect. In this paper we aim to develop a novel framework that could correctly classify textural images with only a small number of training images. By taking into account the repetition and sparsity property of textures we propose a sparse representation based multi-manifold analysis framework for texture classification from few training images. A set of new training samples are generated from each training image by a scale and spatial pyramid, and then the training samples belonging to each class are modelled by a manifold based on sparse representation. We learn a dictionary of sparse representation and a projection matrix for each class and classify the test images based on the projected reconstruction errors. The framework provides a more compact model than the template matching based texture classification methods, and mitigates the overfitting effect. Experimental results show that the proposed method could achieve reasonably high generalization capability even with as few as 3 training images, and significantly outperforms the state-of-the-art texture classification approaches on three benchmark datasets. © 2014 Elsevier B.V. All rights reserved.

Inter-Frame Change Directing online clustering of multiple moving objects for video-based sensor networks

Recognition of multiple moving objects is a very important task for achieving user-cared knowledge to send to the base station in wireless video-based sensor networks. However, video based sensor nodes, which have constrained resources and produce huge amount of video streams continuously, bring a challenge to segment multiple moving objects from the video stream online. Traditional efficient clustering algorithms such as DBSCAN cannot run time-efficiently and even fail to run on limited memory space on sensor nodes, because the number of pixel points is too huge. This paper provides a novel algorithm named Inter-Frame Change Directing Online clustering (IFCDO clustering) for segmenting multiple moving objects from video stream on sensor nodes. IFCDO clustering only needs to group inter-frame different pixels, thus it reduces both space and time complexity while achieves robust clusters the same as DBSCAN. Experiment results show IFCDO clustering excels DBSCAN in terms of both time and space efficiency. © 2008 IEEE.

Recognition as a distinguishing criterion of IS journals

The number of journals publishing information systems (IS) research has grown dramatically over the past few decades. This has resulted in an environment where authors have a wider choice of journals in which to place articles. Electronic journals are now as readily recognised by authorities as print journals. This paper provides firm evidence in support of the assertion that the number of journals publishing IS research has increased. The paper also examines the Australian context where the selection of a journal in which to place an article is influenced by recognition from the Department of Education Science and Training (DEST). In Australia, obtaining DEST recognition as a recognised research journal is not an onerous task, and yet a significant number of IS journals have not done this. Publishing in a DEST recognised journal is essential for Australian researchers to contribute to their organisation’s research quantum and hence research funding. Attention is drawn to an increasing number of IS journals not recognised by DEST, and consequent action is recommended.

Abstracts of the Inaugural Australian Conference for Cognitive Neuropsychology and Cognitive Neuropsychiatry

The first Australian Conference for Cognitive Neuropsychology and Cognitive Neuropsychiatry was conducted by the School of Psychology at Deakin University, Geelong, from Friday, July 13 to Sunday, July 15 and was attended by over 50 cognitive psychologists, neuropsychologists and speech pathologists from Australia, New Zealand, Singapore and the UK. The Conference aimed to bring together researchers from different disciplines including linguistics, psychology, philosophy and speech pathology to present research that relates (neuropsychological or psychiatric) impairment to theories of normal cognitive functioning. The scientific program of the conference included 24 papers of exceptional quality. They were organised into the following thematic sessions: Disorders of language comprehension and production; Semantic memory and category-specific disorders; Reading: development and acquired dyslexia; Writing: development and acquired dyslexia; Memory; Object and face recognition; Theory of mind; Misidentification syndromes. Keynote speakers were Professor Andy Young from the University of York, England and Professor Max Coltheart from the Macquarie Centre for Cognitive Sciences, Sydney.

Classification of face images using local iterated function systems

There has been an increasing interest in face recognition in recent years. Many recognition methods have been developed so far, some very encouraging. A key remaining issue is the existence of variations in the input face image. Today, methods exist that can handle specific image variations. But we are yet to see methods that can be used more effectively in unconstrained situations. This paper presents a method that can handle partial translation, rotation, or scale variations in the input face image. The principal is to automatically identify objects within images using their partial self-similarities. The paper presents two recognition methods which can be used to recognise objects within images. A face recognition system is then presented that is insensitive to limited translation, rotation, or scale variations in the input face image. The performance of the system is evaluated through four experiments. The results show that the system achieves higher recognition rates than those of a number of existing approaches.

Automatic parameters selection for eigenfaces

In this paper, we investigate the parameters selection for Eigenfaces. Our focus is on the eigenvectors and threshold selection issues. We will propose a systematic approach in selecting the eigenvectors based on relative errors of the eigenvalues for the covariance matrix. In addition, we have proposed a method for selecting the classification threshold that utilizes the information obtained from the training data set. Experimentation was conducted on two benchmark face databases, ORL and AMP, with results indicating that the proposed automatic eigenvectors and threshold selection methods produce better recognition performance in terms of precision and recall rates. Furthermore, we show that the eigenvector selection method outperforms energy and stretching dimension methods in terms of selected number of eigenvectors and computation cost.

Exploiting side information in locality preserving projection

Even if the class label information is unknown, side information represents some equivalence constraints between pairs of patterns, indicating whether pairs originate from the same class. Exploiting side information, we develop algorithms to preserve both the intra-class and inter-class local structures. This new type of locality preserving projection (LPP), called LPP with side information (LPPSI), preserves the data's local structure in the sense that the close, similar training patterns will be kept close, whilst the close but dissimilar ones are separated. Our algorithms balance these conflicting requirements, and we further improve this technique using kernel methods. Experiments conducted on popular face databases demonstrate that the proposed algorithm significantly outperforms LPP. Further, we show that the performance of our algorithm with partial side information (that is, using only small amount of pair-wise similarity/dissimilarity information during training) is comparable with that when using full side information. We conclude that exploiting side information by preserving both similar and dissimilar local structures of the data significantly improves performance.

Joint learning and dictionary construction for pattern recognition

We propose a joint representation and classification framework that achieves the dual goal of finding the most discriminative sparse overcomplete encoding and optimal classifier parameters. Formulating an optimization problem that combines the objective function of the classification with the representation error of both labeled and unlabeled data, constrained by sparsity, we propose an algorithm that alternates between solving for subsets of parameters, whilst preserving the sparsity. The method is then evaluated over two important classification problems in computer vision: object categorization of natural images using the Caltech 101 database and face recognition using the Extended Yale B face database. The results show that the proposed method is competitive against other recently proposed sparse overcomplete counterparts and considerably outperforms many recently proposed face recognition techniques when the number training samples is small.

Recognising faces in unseen modes : a tensor based approach

This paper addresses the limitation of current multilinear techniques (multilinear PCA, multilinear ICA) when applied to face recognition for handling faces in unseen illumination and viewpoints. We propose a new recognition method, exploiting the interaction of all the subspaces resulting from multilinear decomposition (for both multilinear PCA and ICA), to produce a new basis called multilinear-eigenmodes. This basis offers the flexibility to handle face images at unseen illumination or viewpoints. Experiments on benchmarked datasets yield superior performance in terms of both accuracy and computational cost.

Boosting performance for 2D linear discriminant analysis via regression

Two Dimensional Linear Discriminant Analysis (2DLDA) has received much interest in recent years. However, 2DLDA could make pairwise distances between any two classes become significantly unbalanced, which may affect its performance. Moreover 2DLDA could also suffer from the small sample size problem. Based on these observations, we propose two novel algorithms called Regularized 2DLDA and Ridge Regression for 2DLDA (RR-2DLDA). Regularized 2DLDA is an extension of 2DLDA with the introduction of a regularization parameter to deal with the small sample size problem. RR-2DLDA integrates ridge regression into Regularized 2DLDA to balance the distances among different classes after the transformation. These proposed algorithms overcome the limitations of 2DLDA and boost recognition accuracy. The experimental results on the Yale, PIE and FERET databases showed that RR-2DLDA is superior not only to 2DLDA but also other state-of-the-art algorithms.