Unconstrained Conditions Research Articles

Learning by observation in the macaque monkey under high experimental constraints

While neuroscience research has tremendously advanced our knowledge about the neural mechanisms of individual learning, i.e. through trial-and-error, it is only recently that neuroscientists have begun to study observational learning, and thus little is known about its neural mechanisms. One limitation is that observational learning has been addressed under unconstrained experimental conditions, not compatible with neuronal recordings. This study examined observational learning in macaque monkeys under the constraining conditions of behavioral neurophysiology. Two animals sat in primate chairs facing each other, with their head fixed. A touch screen was placed face up between the chairs at arm's reach, and the monkeys were trained on an abstract visuomotor associative task. In one experiment, the monkeys alternated the roles of “actor” and “observer”. The actor learned to associate visual cues with reaching targets, while the observer “watched” freely. Then, the observer was given the same cue-target associations just performed by the actor, or had to learn new, not previously observed ones. The results show that learning performance is better after observation. In experiment 2, one monkey learned from a human actor who performed the task with errors only, or with successes only in separate blocks. The monkey's gain in performance was higher after observation of errors than after successes. The findings suggest that observational learning can occur even under highly constraining conditions, and open the way for investigating the neuronal correlates of social learning using the methods of behavioral neurophysiology.

A survey on face detection in the wild: Past, present and future

Face detection is one of the most studied topics in computer vision literature, not only because of the challenging nature of face as an object, but also due to the countless applications that require the application of face detection as a first step. During the past 15years, tremendous progress has been made due to the availability of data in unconstrained capture conditions (so-called ‘in-the-wild’) through the Internet, the effort made by the community to develop publicly available benchmarks, as well as the progress in the development of robust computer vision algorithms. In this paper, we survey the recent advances in real-world face detection techniques, beginning with the seminal Viola–Jones face detector methodology. These techniques are roughly categorized into two general schemes: rigid templates, learned mainly via boosting based methods or by the application of deep neural networks, and deformable models that describe the face by its parts. Representative methods will be described in detail, along with a few additional successful methods that we briefly go through at the end. Finally, we survey the main databases used for the evaluation of face detection algorithms and recent benchmarking efforts, and discuss the future of face detection.

Thermoelectric devices with rotated and coaxial leg configurations: Numerical analysis of performance

More efficient and durable thermoelectric devices are in high demand because of their ability to convert wasted energy into useful electrical energy. Structural integrity must be considered while developing thermoelectric devices because of its impact on power-generation and operation life-time. In this study, the effects of various leg configurations on thermo-mechanical integrity and power-generation performance of thermoelectric devices are investigated. Cylindrical four-leg devices with original-, rotated-, and coaxial-leg configurations are modeled for this purpose. Using these models, finite-element analyses are carried out under thermally steady-state and mechanically unconstrained conditions. The temperature level, the temperature distribution, the developed thermal stresses in the legs, the power output magnitudes, and the conversion efficiencies of the devices are examined individually for a defined temperature gradient. The results indicate that by using a coaxial-leg configuration, maximum thermal stresses in the legs can be reduced 10% while conversion efficiencies can be increased 7%. Models with coaxial-leg configurations also presented slightly different temperature distributions than the other configurations. The effect of rotated-leg configurations on thermal stresses and conversion efficiencies were less than 1.2% and 0.3%, respectively. This study aims to have an impact on considerations of leg configurations for the future of thermoelectric device development.

Facial landmarking for in-the-wild images with local inference based on global appearance

We present a novel method that tackles the problem of facial landmarking in unconstrained conditions within the part-based framework. Part-based methods alternate the evaluation of local appearance models to produce a per-point response map and a shape fitting step which finds a valid face shape that maximises the sum of the per-point responses. Our approach focuses on obtaining better appearance models for the creation of the response maps, and it can be used in combination with any shape fitting strategy. Local appearance models need to tackle very heterogeneous data when dealing with in-the-wild imagery due to factors as varying head poses, facial expressions, identity, lighting conditions, or image quality among others. Pose-wise experts are typically used in this scenario so that each expert deals with more homogeneous data. However, the computation cost at test time is significantly increased. Furthermore, choosing the right expert is not straightforward, which can lead to gross errors. We propose to dynamically select at test time the training examples used for inference. We use a global similarity measure to select the most adequate training examples for inference, and create a single test sample-specific expert using a localised inference technique. To illustrate the validity of these ideas, we capitalise on the recently proposed use of regression to generate local appearance models. In particular, we use Gaussian processes, as their non-parametric nature easily allows for localised regression. This novel way of constructing the response maps is combined with two state-of-the-art standard shape fitting algorithms, the popular Constrained Local Models framework and the Consensus of Exemplars method. We validate our method on two publicly available datasets as well as on a cross-dataset experiment, showing a considerable performance improvement of the proposed approach.

Sensor positioning and experimental constraints influence estimates of local dynamic stability during repetitive spine movements

Application of non-linear dynamics analyses to study human movement has increased recently, which necessitates an understanding of how dependent measures may be influenced by experimental design and setup. Quantifying local dynamic stability for a multi-articulated structure such as the spine presents the possibility for estimates to be influenced by positioning of kinematic sensors used to measure spine angular kinematics. Oftentimes researchers will also choose to constrain the spine׳s movement by physically restraining the pelvis and/or using targets to control movement endpoints. Ten healthy participants were recruited, and asked to perform separate trials of 35 consecutive cycles of spine flexion under both constrained and unconstrained conditions. Electromagnetic sensors that measure three-dimensional angular orientations were positioned over the pelvis and the spinous processes of L3, L1, and T11. Using the pelvic sensor as a reference, each sensor location on the spine was used to obtain a different representation of the three-dimensional spine angular kinematics. Local dynamic stability of each kinematic time-series was determined by calculating the maximum finite-time Lyapunov exponent (λmax). Estimates for λmax were significantly lower (i.e. dynamically more stable) for spine kinematic data obtained from the L3 sensor than those obtained from kinematic data using either the L1 or T11 sensors. Likewise, λmax was lower when the movement was constrained. These results emphasize the importance of proper placement of instrumentation for quantifying local dynamic stability of spine kinematics and are especially relevant for repeated measures designs where data are obtained from the same individual on multiple days.

Association and causation in epidemiology - half a century since the publication of Bradford Hill's interpretational guidance.

Austin Bradford Hill was one of the greats in the fields of epidemiology and medical statistics. In the mid-20th century, with another great, Richard Doll, Bradford Hill initiated epidemiological studies that were to be highly influential in revealing the causal link between cigarette smoking and lung cancer. However, this link was not accepted without a battle, and opponents of a direct cause-and-effect interpretation of the epidemiological association included such notables as the eminent statistician Ronald Fisher. The debate spurred Bradford Hill to consider in some depth how the findings of epidemiological studies should be interpreted, and this led to the publication in 1965 in Proceedings of the Royal Society of Medicine of his seminal paper on association and causation. To mark the 50th anniversary of the publication of this landmark paper, it is reproduced in this issue of Journal of the Royal Society of Medicine. Bradford Hill’s 1965 paper is a remarkable one that is full of insights. It proposes nine guidelines (often erroneously referred to as ‘criteria’, which Bradford Hill made clear they were not) against which a statistical association found in an epidemiological study may be judged as to whether a causal interpretation is reasonable or not. The most important of these guidelines are ‘strength’ (a strong association is more likely to be causal than a weak one), ‘consistency’ (an association is observed in different studies, under different circumstances, times and places), ‘biological gradient’ (i.e. doseresponse – the effect should tend to be greater with a higher level of exposure) and ‘temporality’ (the effect follows the potential cause after an appropriate interval). Another guideline, ‘biological plausibility’, increases in importance as fundamental knowledge of disease aetiology accumulates, but such knowledge is clearly not complete, which should introduce a note of caution in placing too much emphasis on this guideline; nonetheless, the direct causal interpretation of certain statistical associations would stretch credibility given what is known today. What is it about epidemiology that demands a detailed examination of the interpretation of its findings such as that conducted by Bradford Hill? With the exception of a few studies that are able to ‘piggyback’ on clinical trials, epidemiology is an observational (i.e. non-experimental) science. This is in contrast to randomised controlled clinical trials, which are experimental set-ups: study subjects are randomised between treatment groups by the investigators so that any background individual differences that might affect the outcome of the trial, even if unknown, are ‘evened out’ between the groups. (Randomisation between treatment groups is a fundamental concept in clinical trials that Bradford Hill was key in establishing as a necessary requirement in study design.) In observational epidemiology, randomisation of study subjects between groups (say, different levels of exposure to tobacco smoke) is not possible because studies in which people are deliberately exposed to potentially harmful substances without a realistic prospect of personal benefit are unethical. As a consequence, epidemiology must rely on data generated under the unconstrained conditions of everyday life, with no intervention on the part of the investigators, and this greatly complicates the interpretation of epidemiological findings. In clinical trials, essentially all that needs to be considered (assuming the study has been correctly designed, conducted and analysed) in judging causality is whether chance is a reasonable alternative explanation for the findings, i.e. how statistically significant are the results. In epidemiology, not only must statistical fluctuations be taken into account but also the potential presence of systematic errors (of primary concern is the existence of bias in study data, but other errors are possible) and confounding (when a factor considered in a study is associated with another factor that influences the outcome, producing a distorted, potentially misleading, result).

Multimodal biometric system built on the new entropy function for feature extraction and the Refined Scores as a classifier

This paper presents a unique face based multimodal biometric system comprising IR face, ear and iris to cater to the surveillance application by proposing new entropy function. Two new features based on this entropy are devised to cater the highly uncertain database found at the surveillance site. To handle the erroneous scores we have proposed Refined Score (RS) method and applied it on individual IR face, ear and iris modalities under both constrained and the unconstrained conditions for the authentication of users and also used for the score level fusion of these modalities using the proposed entropy based features. The entropy features show good performance under the constrained and unconstrained databases whereas the conventional entropies do not fare well on the unconstrained databases. RS based classifier always outperforms the EC (Euclidean classifier) and RS based score level fusion has an edge over the conventional score level fusion.

Histogram of gradient phases: a new local descriptor for face recognition

Gradient‐based local descriptors have received more attention these years and have been successfully used in many applications such as human detection and face recognition. The advantages of the local descriptors are the resistance to the local geometric and photometric errors and the robustness to the expression variations. In this paper, the authors propose a new local descriptor called the histogram of gradient phases (HGP), which has some intriguing properties compared with the existing local descriptors such as the histogram of orientated gradients and DAISY for face recognition under the unconstrained conditions. In contrast with the histogram of the oriented gradient descriptor, the orientation histogram is computed from the estimated gradient phase distributions instead of weighting the votes of the gradient magnitudes. In this paper, the phase distributions are estimated by means of the gradient phases and the variances are decided by the estimated gradient signal‐to‐noise ratios of the pixels in a local region. The HGP descriptor which takes the confidence of the gradient phase into account is more discriminative and less sensitive to the normalisation process than most local descriptors, which significantly degrade without a proper normalisation. The simulation results show that the proposed HGP descriptor achieves a better performance and is more robust than the existing local descriptors.

A robust two‐stage face recognition system with localisation error compensation

In practical systems, face recognition under unconstrained conditions is a very challenging task, where their input images are first pre‐processed and initially aligned by a face detection algorithm. However, there are still some residual localisation errors after the initial alignment. If we do not take these errors into account, the recognition performance should be greatly degraded for most face recognition algorithms. Generally, when designing a practical face recognition system, we need to compromise the capability of residual error tolerance and the discriminating capability. Although it is feasible to apply an iterative alignment algorithm to fine‐tune alignment, it will increase the computation load significantly. In this study, we propose an adaptive two‐stage face recognition system consisting of two block‐based recognition stages with a relatively larger cell size (i.e. the size of local regions) in the first stage to provide sufficient tolerance for geometric errors followed by a smaller one in the second stage to accurately evaluate a most probable candidate subset, which is adaptively determined according to the proposed confidence measure. In addition, an iterative gradient‐based alignment algorithm is incorporated into the two‐stage system to refine the alignment such that the recognition performance can be improved and the computation load can be saved simultaneously.

Active Orientation Models for Face Alignment In-the-Wild

We present Active Orientation Models (AOMs), generative models of facial shape and appearance, which extend the well-known paradigm of Active Appearance Models (AAMs) for the case of generic face alignment under unconstrained conditions. Robustness stems from the fact that the proposed AOMs employ a statistically robust appearance model based on the principal components of image gradient orientations. We show that when incorporated within standard optimization frameworks for AAM learning and fitting, this kernel Principal Component Analysis results in robust algorithms for model fitting. At the same time, the resulting optimization problems maintain the same computational cost. As a result, the main similarity of AOMs with AAMs is the computational complexity. In particular, the project-out version of AOMs is as computationally efficient as the standard project-out inverse compositional algorithm, which is admittedly one of the fastest algorithms for fitting AAMs. We verify experimentally that: 1) AOMs generalize well to unseen variations and 2) outperform all other state-of-the-art AAM methods considered by a large margin. This performance improvement brings AOMs at least in par with other contemporary methods for face alignment. Finally, we provide MATLAB code at http://ibug.doc.ic.ac.uk/resources.

A kernelized sparsity-based approach for best spectral bands selection for face recognition

We study face recognition in unconstrained illumination conditions. A twofold contribution is proposed: First, the robustness of four state-of-the-art algorithms, namely Multi-block Local Binary Pattern (MBLBP), Histogram of Gabor Phase Patterns (HGPP), Local Gabor Binary Pattern Histogram Sequence (LGBPHS) and Patterns of Oriented Edge Magnitudes (POEM-WPCA) against high illumination variation is studied. Second, we propose to enhance the performance of the four mentioned algorithms, which has been drastically decreased upon the day lighted face images provided by IRIS-M3 face database. For this purpose, we use visible narrow band subspectral images selected from the mentioned database. We formulate best spectral bands selection as a pursuit optimization problem wherein the vector of weights determining the importance of each visible spectral band is supposed to be sparse, and hence can be determined by minimizing its L1-norm. Several fusing approaches are then applied on selected best spectral bands using multi-scale and multi-orientation Gabor wavelets. The results highlight further the still challenging problem of face recognition in conditions with high illumination variation, as well as the effectiveness of our subspectral images based approach with its two components; bands selection and bands fusion, to increase the accuracy of the studied algorithms by at least 14 % upon the proposed database.

Skin color modeling for face detection and segmentation: a review and a new approach

Detecting skin-colored pixels, although seems a straightforward easy task, has proven to be quite a challenging task in images that are captured under complex unconstrained imaging conditions. Color segmentation is an important method of segmenting the body in images. In this paper, we first provides a review of skin color modeling research works for face detection. In particular, we focus on discussing the challenges of skin color modeling, detection, and segmentation, and then we present the skin color modeling and region-based skin segmentation. We also propose a new approach to analyze the hue characteristics of monitoring personnel in coal mine. We determine the skin color boundaries of the miners using the boundary detection method based on the response of visual hue. And then, we apply to determining hue area method using dual hue detector to segment skin color for miners. Experiments show that the method has excellent ability to segment skin regions of miners. We also summarize the most widely used methods for skin detection approaches and collecting their numerical evaluation results.

Contour Model-Based Hand-Gesture Recognition Using the Kinect Sensor

In RGB-D sensor-based pose estimation, training data collection is often a challenging task. In this paper, we propose a new hand motion capture procedure for establishing the real gesture data set. A 14-patch hand partition scheme is designed for color-based semiautomatic labeling. This method is integrated into a vision-based hand gesture recognition framework for developing desktop applications. We use the Kinect sensor to achieve more reliable and accurate tracking under unconstrained conditions. Moreover, a hand contour model is proposed to simplify the gesture matching process, which can reduce the computational complexity of gesture matching. This framework allows tracking hand gestures in 3-D space and matching gestures with simple contour model, and thus supports complex real-time interactions. The experimental evaluations and a real-world demo of hand gesture interaction demonstrate the effectiveness of this framework.

Smartphone based visible iris recognition using deep sparse filtering

Good biometric performance of iris recognition motivates it to be used for many large scale security and access control applications. Recent works have identified visible spectrum iris recognition as a viable option with considerable performance. Key advantages of visible spectrum iris recognition include the possibility of iris imaging in on-the-move and at-a-distance scenarios as compared to fixed range imaging in near-infra-red light. The unconstrained iris imaging captures the images with largely varying radius of iris and pupil. In this work, we propose a new segmentation scheme and adapt it to smartphone based visible iris images for approximating the radius of the iris to achieve robust segmentation. The proposed technique has shown the improved segmentation accuracy up to 85% with standard OSIRIS v4.1. This work also proposes a new feature extraction method based on deep sparse filtering to obtain robust features for unconstrained iris images. To evaluate the proposed segmentation scheme and feature extraction scheme, we employ a publicly available database and also compose a new iris image database. The newly composed iris image database (VSSIRIS) is acquired using two different smartphones – iPhone 5S and Nokia Lumia 1020 under mixed illumination with unconstrained conditions in visible spectrum. The biometric performance is benchmarked based on the equal error rate (EER) obtained from various state-of-art schemes and proposed feature extraction scheme. An impressive EER of 1.62% is obtained on our VSSIRIS database and an average gain of around 2% in EER is obtained on the public database as compared to the well-known state-of-art schemes.

A deep structure for human pose estimation

Articulated human pose estimation in unconstrained conditions is a great challenge. We propose a deep structure that represents a human body in different granularity from coarse-to-fine for better detecting parts and describing spatial constrains between different parts. Typical approaches for this problem just utilize a single level structure, which is difficult to capture various body appearances and hard to model high-order part dependencies. In this paper, we build a three layer Markov network to model the body structure that separates the whole body to poselets (combined parts) then to parts representing joints. Parts at different levels are connected through a parent–child relationship to represent high-order spatial relationships. Unlike other multi-layer models, our approach explores more reasonable granularity for part detection and sophisticatedly designs part connections to model body configurations more effectively. Moreover, each part in our model contains different types so as to capture a wide range of pose modes. And our model is a tree structure, which can be trained jointly and favors exact inference. Extensive experimental results on two challenging datasets show the performance of our model improving or being on-par with state-of-the-art approaches.

Compressibility of sorptive porous media: Part 2. Experimental study on coal

This paper, second of a two-part series, discusses the results of the experimental work conducted to estimate the different compressibilities of coal under both unconstrained and constrained conditions. Under unconstrained conditions, the shrinkage or swelling compressibility (C m) was measured, which was found with certainty to be a pressure-dependent parameter. The model proposed to estimate was able to effectively predict the variation trend, although the modeled values were larger than those calculated using experimental results. The pore-volume compressibility under uniaxial strain conditions for helium depletion was found to be a constant positive value. The value of for methane depletion, however, was found negative, indicating that the pore volume (cleat) increases with depletion. Moreover, its absolute value increased with decreasing methane pressure. Consistent with field permeability observations, the permeability increases with methane depletion, and the rate of increase at lower pressures is higher than at high pressures. The proposed pore-volume compressibility model was well able to predict the variation.

A new entropy function and a classifier for thermal face recognition

An attempt is made to devise a new entropy function that goes beyond the existing entropy functions with its ability to change the information source values (gray levels in an IR image) and its information gain by selecting its parameters. Our objective is to improve the existing results on the Infra-Red thermal face recognition by using this entropy function that possesses peculiar characteristics such as splitting and inverting which impart a discriminating power. To cash on its discriminating power, two types of features Effective Gaussian Information (EGI) source and Effective Exponential Information (EEI) source functions are developed. To classify the features, we have modified our earlier classifier (Mamta and Hanmandlu, 2014) using the new entropy function. The performance of the new features and new classifier is tested on IR face databases under the constrained and the unconstrained conditions with regard to occlusion, noise and low resolution. A comparison of performance shows that the new entropy function outperforms the existing entropy functions such as Shannon, Renyi, Tsallis and Pal and Pal, Collision, Min entropy and Susan–Hanman.

Evaluación del impacto del preprocesamiento de imágenes en la segmentación del iris

La segmentación del iris es una de las etapas más importantes en los sistemas de reconocimiento del iris. En este trabajo se aplican algoritmos de preprocesamiento de la imagen con el objetivo de evaluar su impacto en los porcentajes de segmentación exitosa del iris. Los algoritmos utilizados se basan en el ajuste del histograma, filtros Gaussianos y en la eliminación del reflejo especular en imágenes del ojo humano. Se aplica el método de segmentación introducido por Masek a 199 imágenes tomadas bajo condiciones no controladas, pertenecientes a la base de datos CASIA-irisV3, antes y después de aplicar los algoritmos de preprocesamiento. Posteriormente se evalúa el impacto de los algoritmos de preprocesamiento en el porcentaje de segmentación exitosa del iris por medio de una inspección visual de las imágenes, para determinar si las circunferencias detectadas del iris y de la pupila corresponden adecuadamente con el iris y la pupila de la imagen real. El algoritmo que generó uno de los mayores incrementos de los porcentajes de segmentación exitosa (pasa de 59% a 73%) es aquel que combina la eliminación de reflejos especulares, seguido por la aplicación de un filtro Gaussiano con máscara 5x5. Los resultados obtenidos señalan la importancia de una etapa previa de preprocesamiento de la imagen como paso previo para garantizar una mayor efectividad en el proceso de detección de bordes y segmentación del iris.

Towards an automated acoustic detection system for free-ranging elephants

The human–elephant conflict is one of the most serious conservation problems in Asia and Africa today. The involuntary confrontation of humans and elephants claims the lives of many animals and humans every year. A promising approach to alleviate this conflict is the development of an acoustic early warning system. Such a system requires the robust automated detection of elephant vocalizations under unconstrained field conditions. Today, no system exists that fulfils these requirements. In this paper, we present a method for the automated detection of elephant vocalizations that is robust to the diverse noise sources present in the field. We evaluate the method on a data-set recorded under natural field conditions to simulate a real-world scenario. The proposed method outperformed existing approaches and robustly and accurately detected elephants. It thus can form the basis for a future automated early warning system for elephants. Furthermore, the method may be a useful tool for scientists in bioacoustics for the study of wildlife recordings.

Robust authentication using the unconstrained infrared face images

Face recognition under the unconstrained conditions that exist in surveillance is the need of the present times. Thus for high end security the research on IR based face recognition assumes importance because of its insensitivity to illumination, disguise and surgery. This paper presents IR face based biometric authentication using the information-set based four types of interactive features and two classifiers. The information sets originate from a fuzzy set on representing the uncertainty associated with the information source instead of a membership function which gives only the degree of association to the fuzzy set. The four feature types include the effective exponential information source (EEI), the effective Gaussian information source (EGI), the effective multi quadratic information source (EMQDI) and inverse of this feature (EIMQDI). The interactive features are obtained by taking the s-norms on the features from the successive windows. Two classifiers called the Hanman Classifier and the weighted Hanman Classifier are formulated using t-norms. The features and classifiers are tested on the created databases incorporating the unconstrained conditions such as occlusion, less resolution and noise.