Incremental learning of temporally-coherent gaussian mixture models

In this paper we address the problem of learning Gaussian Mixture Models (GMMs) incrementally. Unlike previous approaches which universally assume that new data comes in blocks representable by GMMs which are then merged with the current model estimate, our method works for the case when novel data points arrive oneby- one, while requiring little additional memory. We keep only two GMMs in the memory and no historical data. The current fit is updated with the assumption that the number of components is fixed, which is increased (or reduced) when enough evidence for a new component is seen. This is deduced from the change from the oldest fit of the same complexity, termed the Historical GMM, the concept of which is central to our method. The performance of the proposed method is demonstrated qualitatively and quantitatively on several synthetic data sets and video sequences of faces acquired in realistic imaging conditions

Incremental learning of temporally-coherent Gaussian mixture models

In this paper we address the problem of learning Gaussian Mixture Models (GMMs) incrementally. Unlike previous approaches which universally assume that new data comes in blocks representable by GMMs which are then merged with the current model estimate, our method works for the case when novel data points arrive one- by-one, while requiring little additional memory. We keep only two GMMs in the memory and no historical data. The current fit is updated with the assumption that the number of components is fixed which is increased (or reduced) when enough evidence for a new component is seen. This is deducedfrom the change from the oldest fit of the same complexity, termed the Historical GMM, the concept of which is central to our method. The performance of the proposed method is demonstrated qualitatively and quantitatively on several synthetic data sets and video sequences of faces acquired in realistic imaging conditions.

Dirichlet process mixture models with pairwise constraints for data clustering

The Dirichlet process mixture (DPM) model, a typical Bayesian nonparametric model, can infer the number of clusters automatically, and thus performing priority in data clustering. This paper investigates the influence of pairwise constraints in the DPM model. The pairwise constraint, known as two types: must-link (ML) and cannot-link (CL) constraints, indicates the relationship between two data points. We have proposed two relevant models which incorporate pairwise constraints: the constrained DPM (C-DPM) and the constrained DPM with selected constraints (SC-DPM). In C-DPM, the concept of chunklet is introduced. ML constraints are compiled into chunklets and CL constraints exist between chunklets. We derive the Gibbs sampling of the C-DPM based on chunklets. We further propose a principled approach to select the most useful constraints, which will be incorporated into the SC-DPM. We evaluate the proposed models based on three real datasets: 20 Newsgroups dataset, NUS-WIDE image dataset and Facebook comments datasets we collected by ourselves. Our SC-DPM performs priority in data clustering. In addition, our SC-DPM can be potentially used for short-text clustering.

Unifying background models over complex audio using entropy

In this paper we extend an existing audio background modelling technique, leading to a more robust application to complex audio environments. The determination of background audio is used as an initial stage in the analysis of audio for surveillance and monitoring applications. Knowledge of the background serves to highlight unusual or infrequent sounds. An existing modelling approach uses an online, adaptive Gaussian Mixture model technique that uses multiple distributions to model variations in the background. The method used to determine the background distributions of the GMM leads to a failure mode of the existing technique when applied to complex audio. We propose a method incorporating further information, the proximity of distributions determined using entropy, to determine a more complete background model. The method was successful in more robustly modelling the background for complex audio scenes.

Accurate and efficient face recognition from video

As a problem of high practical appeal but outstanding challenges, computer-based face recognition remains a topic of extensive research attention. In this paper we are specifically interested in the task of identifying a person from multiple training and query images. Thus, a novel method is proposed which advances the state-of-the-art in set based face recognition. Our method is based on a previously described invariant in the form of generic shape-illumination effects. The contributions include: (i) an analysis of computational demands of the original method and a demonstration of its practical limitations, (ii) a novel representation of personal appearance in the form of linked mixture models in image and pose-signature spaces, and (iii) an efficient (in terms of storage needs and matching time) manifold re-illumination algorithm based on the aforementioned representation. An evaluation and comparison of the proposed method with the original generic shape-illumination algorithm shows that comparably high recognition rates are achieved on a large data set (1.5% error on 700 face sets containing 100 individuals and extreme illumination variation) with a dramatic improvement in matching speed (over 700 times for sets containing 1600 faces) and storage requirements (independent of the number of training images).

Hallucinating optimal high-dimensional subspaces

Linear subspace representations of appearance variation are pervasive in computer vision. This paper addresses the problem of robustly matching such subspaces (computing the similarity between them) when they are used to describe the scope of variations within sets of images of different (possibly greatly so) scales. A naïve solution of projecting the low-scale subspace into the high-scale image space is described first and subsequently shown to be inadequate, especially at large scale discrepancies. A successful approach is proposed instead. It consists of (i) an interpolated projection of the low-scale subspace into the high-scale space, which is followed by (ii) a rotation of this initial estimate within the bounds of the imposed "downsampling constraint". The optimal rotation is found in the closed-form which best aligns the high-scale reconstruction of the low-scale subspace with the reference it is compared to. The method is evaluated on the problem of matching sets of (i) face appearances under varying illumination and (ii) object appearances under varying viewpoint, using two large data sets. In comparison to the naïve matching, the proposed algorithm is shown to greatly increase the separation of between-class and within-class similarities, as well as produce far more meaningful modes of common appearance on which the match score is based. © 2014 Elsevier Ltd.

Fast road detection and tracking in aerial videos

 We propose a fast approach for detecting and tracking a specific road in aerial videos. It combines adaptive Gaussian Mixture Models (GMMs) to describe road colour distributions, and homography based tracking to track road geometries, where an efficient technique is developed to estimate homography transformations between two frames. Experiments are conducted on videos captured by our unmanned aerial vehicles. All the results demonstrate the effectiveness of our proposed method. We test 1755 frames from 5 videos. Our approach can achieve 0.032 seconds per frame and 2.64% segmentation error for images with 908 × 513 resolutions, on average.

Bayesian nonparametric multilevel clustering with group-level contexts

We present a Bayesian nonparametric framework for multilevel clustering which utilizes group- level context information to simultaneously discover low-dimensional structures of the group contents and partitions groups into clusters. Using the Dirichlet process as the building block, our model constructs a product base-measure with a nested structure to accommodate content and context observations at multiple levels. The proposed model possesses properties that link the nested Dinchiet processes (nDP) and the Dirichlet process mixture models (DPM) in an interesting way: integrating out all contents results in the DPM over contexts, whereas integrating out group-specific contexts results in the nDP mixture over content variables. We provide a Polyaurn view of the model and an efficient collapsed Gibbs inference procedure. Extensive experiments on real-world datasets demonstrate the advantage of utilizing context information via our model in both text and image domains.

A summative scoring system for evaluation of human kinematic performance

Evaluation of human kinematic performance is essential in rehabilitation and skill assessment. These services are in high demand where the improvements made due to exercises need to be regularly assessed. In some relevant industries there is a need to evaluate their employee capabilities quantitatively for accident compensation and insurance purposes. In particular, these assessments are preferred to be based on more quantifiable measures in a standardized form ensuring accuracy, reliability, ease of use and anywhere anytime information to the clinician. Therefore, it is necessary to have an efficient mechanism for evaluation and assessment of human kinematic movements as the current motion matching and recognition algorithms fall short due to characteristically strict specifications required in numerous health care applications. In this paper, we propose a summative approach using a double integral to define a closeness between two trajectories typically generated by human movement. This approach can be considered as a spatial scoring mechanism in the evaluation of human kinematic performance as well as in movement recognition applications. Several experiments based on computer simulations as well as real data were set up to examine the performance of the proposed approach as a scoring mechanism for the evaluation of human kinematic performances. The results demonstrated better characterization of the movement assessment and motion recognition ability, with a recognition rate of 86.19%, than the currently used methods such as Gaussian mixture models and pose normalization employed in motion recognition tasks. Finally, we use the scoring mechanism to analyze the proximity in human kinematic performance.

Spatial dynamics of the knob-tailed gecko Nephrurus stellatus in a fragmented agricultural landscape

In fragmented landscapes, a species' dispersal ability and response to habitat condition are key determinants of persistence. To understand the relative importance of dispersal and condition for survival of Nephrurus stellatus (Gekkonidae) in southern Australia, we surveyed 92 woodland remnants three times. This gecko favours early post-fire succession conditions so may be at risk of extinction in the long-unburnt agricultural landscape. Using N-mixture models, we compared the influence of four measures of isolation, patch area and two habitat variables on the abundance and occurrence of N. stellatus, while taking into account detection probability. Patch occupancy was high, despite the long-term absence of fire from most remnants. Distance to the nearest occupied site was the most informative measure of patch isolation, exhibiting a negative relationship with occupancy. Distance to a nearby conservation park had little influence, suggesting that mainland-island metapopulation dynamics are not important. Abundance and occurrence were positively related to %-cover of spinifex (Triodia), indicating that niche-related factors may also contribute to spatial dynamics. Patterns of patch occupancy imply that N. stellatus has a sequence of spatial dynamics across an isolation gradient, with patchy populations and source-sink dynamics when patches are within 300 m, metapopulations at intermediate isolation, and declining populations when patches are separated by >1-2 km. Considering the conservation needs of the community, habitat condition and connectivity may need to be improved before fire can be reintroduced to the landscape. We speculate that fire may interact with habitat degradation and isolation, increasing the risk of local extinctions.

Solving regression problems using competitive ensemble models

The use of ensemble models in many problem domains has increased significantly in the last fewyears. The ensemble modeling, in particularly boosting, has shown a great promise in improving predictive performance of a model. Combining the ensemble members is normally done in a co-operative fashion where each of the ensemble members performs the same task and their predictions are aggregated to obtain the improved performance. However, it is also possible to combine the ensemble members in a competitive fashion where the best prediction of a relevant ensemble member is selected for a particular input. This option has been previously somewhat overlooked. The aim of this article is to investigate and compare the competitive and co-operative approaches to combining the models in the ensemble. A comparison is made between a competitive ensemble model and that of MARS with bagging, mixture of experts, hierarchical mixture of experts and a neural network ensemble over several public domain regression problems that have a high degree of nonlinearity and noise. The empirical results showa substantial advantage of competitive learning versus the co-operative learning for all the regression problems investigated. The requirements for creating the efficient ensembles and the available guidelines are also discussed.

New village models : examining the sustainability of existing realities and hypothetical visions

Zavoj is a mountain village in the Republic of Macedonia. While the village has become a site of emigration due to the mass exodus of its inhabitants, new house constructions have appeared. Scattered through deteriorating vernacular dwellings are new houses, and new fragments of houses attached to old houses. These new houses and house-fragments are introduced by emigrants returning to the village temporarily, and not by the remaining local village inhabitants. In the new millennium their number has dramatically increased. Migration has produced an incongruous mixture of architectures giving rise to questions about sustainable development in relation to new constructions in traditional environments? New buildings in the village are symptomatic of a much more universal phenomenon that is transforming vast rural landscapes into loosely urbanized regions. In contrast to this existing reality, a program in the Faculty of Architecture, Ss. Kiril & Metodij University, Skopje, Republic of Macedonia, is exploring alternative visions for the revitalization of the village. Can these offer more sustainable design approaches to the village? This paper examines sustainability in the dialectic between new constructions in existing villages and hypothetical visions for the new village.

Generative models for automatic recognition of human daily activities from a single triaxial accelerometer

In this work, we compare two generative models including Gaussian Mixture Model (GMM) and Hidden Markov Model (HMM) with Support Vector Machine (SVM) classifier for the recognition of six human daily activity (i.e., standing, walking, running, jumping, falling, sitting-down) from a single waist-worn tri-axial accelerometer signals through 4-fold cross-validation and testing on a total of thirteen subjects, achieving an average recognition accuracy of 96.43% and 98.21% in the first experiment and 95.51% and 98.72% in the second, respectively. The results demonstrate that both HMM and GMM are not only able to learn but also capable of generalization while the former outperformed the latter in the recognition of daily activities from a single waist worn tri-axial accelerometer. In addition, these two generative models enable the assessment of human activities based on acceleration signals with varying lengths.

Small-variance asymptotics for bayesian nonparametric models with constraints

The users often have additional knowledge when Bayesian nonparametric models (BNP) are employed, e.g. for clustering there may be prior knowledge that some of the data instances should be in the same cluster (must-link constraint) or in different clusters (cannot-link constraint), and similarly for topic modeling some words should be grouped together or separately because of an underlying semantic. This can be achieved by imposing appropriate sampling probabilities based on such constraints. However, the traditional inference technique of BNP models via Gibbs sampling is time consuming and is not scalable for large data. Variational approximations are faster but many times they do not offer good solutions. Addressing this we present a small-variance asymptotic analysis of the MAP estimates of BNP models with constraints. We derive the objective function for Dirichlet process mixture model with constraints and devise a simple and efficient K-means type algorithm. We further extend the small-variance analysis to hierarchical BNP models with constraints and devise a similar simple objective function. Experiments on synthetic and real data sets demonstrate the efficiency and effectiveness of our algorithms.