Global Gaussian Mixture Model Research Articles

Federated Data Quality Assessment Approach: Robust Learning With Mixed Label Noise.

Federated learning (FL) has been an effective way to train a machine learning model distributedly, holding local data without exchanging them. However, due to the inaccessibility of local data, FL with label noise would be more challenging. Most existing methods assume only open-set or closed-set noise and correspondingly propose filtering or correction solutions, ignoring that label noise can be mixed in real-world scenarios. In this article, we propose a novel FL method to discriminate the type of noise and make the FL mixed noise-robust, named FedMIN. FedMIN employs a composite framework that captures local-global differences in multiparticipant distributions to model generalized noise patterns. By determining adaptive thresholds for identifying mixed label noise in each client and assigning appropriate weights during model aggregation, FedMIN enhances the performance of the global model. Furthermore, FedMIN incorporates a loss alignment mechanism using local and global Gaussian mixture models (GMMs) to mitigate the risk of revealing samplewise loss. Extensive experiments are conducted on several public datasets, which include the simulated FL testbeds, i.e., CIFAR-10, CIFAR-100, and SVHN, and the real-world ones, i.e., Camelyon17 and multiorgan nuclei challenge (MoNuSAC). Compared to FL benchmarks, FedMIN improves model accuracy by up to 9.9% due to its superior noise estimation capabilities.

High Speed Image Searching For Human Gait Feature Selection

In this paper, a new patch distribution feature (i.e., referred to as Gabor-PDF) is used for human gait recognition. We represent each gait energy image (GEI) as a set of local augmented Gabor features, which concatenate the Gabor features extracted from different scales and different orientations. A global Gaussian mixture model (GMM) (i.e., referred to as the universal back- ground model) with the local augmented Gabor features from all the gallery GEIs; then, each gallery or probe GEI is further expressed as the normalized parameters of an image-specific GMM adapted from the global GMM. Observing that one video is naturally represented as a group of GEIs, also a new classification method called locality-constrained group sparse representation (LGSR) to classify each probe video by minimizing the weighted l1,2 mixed-norm-regularized reconstruction error with respect to the gallery videos. In contrast to the standard group sparse representation method that is a special case of LGSR, the group sparsity and local smooth sparsity constraints are both enforced in LGSR. The same LGSR algorithm is used for both color images and for content based image retrieval (CBIR

Distributed Gaussian mixture model for monitoring plant-wide processes with multiple operating modes

For large-scale plant-wide processes with multiple operating conditions, a distributed Gaussian mixture modeling and monitoring mechanism is proposed. To overcome the deficient prior modeling knowledge for complex process, a two-dimensional probabilistic topic model based technique named Bayesian co-clustering method is developed to simultaneously conduct the sub-block division and operating mode recognition. With the obtained sub-blocks and operating modes, a global Gaussian mixture model is first built and then several local Gaussian mixture models are extracted and applied for distributed monitoring of plant-wide processes. By conducting distributed modeling and monitoring, both global and local process changes can be reflected and the fault region can be localized more easily for further analyses. The feasibility and effectiveness of the proposed method is confirmed through a numerical example and the Tennessee Eastman benchmark process.

Efficient visual object detection with spatially global Gaussian mixture models and uncertainties

In this paper, we deal with the problem of visual detection of moving objects using innovative Gaussian mixture models (GMM). The proposed method, the Spatially Global Gaussian Mixture Model (SGGMM) uses RGB and pixel uncertainty for background modelling. The SGGMM with colours only is used for scenes with moderate illumination changes. When combined with pixel uncertainty statistics, the method can deal efficiently with dynamic backgrounds and scene backgrounds with fast change in luminosity. Experimental evaluation in indoor and outdoor environments shows the performance of the foreground segmentation with the proposed SGGMM model using solely RGB colour or in combination with pixel uncertainties. These experimental scenarios take into account changes in the background within the scene. They are also used to compare the proposed technique with other state-of-the-art segmentation approaches in terms of accuracy and execution time performance. Further, our solution is implemented and tested in embedded camera sensor network nodes.

Action recognition using lie algebrized gaussians over dense local spatio-temporal features

This paper presents a novel framework for human action recognition based on a newly proposed mid-level feature representation method named Lie Algebrized Guassians (LAG). As an action sequence can be treated as a 3D object in space-time space, we address the action recognition problem by recognizing 3D objects and characterize 3D objects by the probability distributions of local spatio-temporal features. First, for each video, we densely sample local spatio-temporal features (e.g. HOG3D) at multiple scales confined in bounding boxes of human body. Moreover, normalized spatial coordinates are appended to local descriptor in order to capture spatial position information. Then the distribution of local features in each video is modeled by a Gaussian Mixture Model (GMM). To estimate the parameters of video-specific GMMs, a global GMM is trained using all training data and video-specific GMMs are adapted from the global GMM. Then the LAG is adopted to vectorize those video-specific GMMs. Finally, linear SVM is employed for classification. Experimental results on the KTH and UCF Sports dataset show that our method achieves state-of-the-art performance.

Action Recognition Using Multilevel Features and Latent Structural SVM

We first propose a new low-level visual feature, called spatio-temporal context distribution feature of interest points, to describe human actions. Each action video is expressed as a set of relative XYT coordinates between pairwise interest points in a local region. We learn a global Gaussian mixture model (GMM) (referred to as a universal background model) using the relative coordinate features from all the training videos, and then we represent each video as the normalized parameters of a video-specific GMM adapted from the global GMM. In order to capture the spatio-temporal relationships at different levels, multiple GMMs are utilized to describe the context distributions of interest points over multiscale local regions. Motivated by the observation that some actions share similar motion patterns, we additionally propose a novel mid-level class correlation feature to capture the semantic correlations between different action classes. Each input action video is represented by a set of decision values obtained from the pre-learned classifiers of all the action classes, with each decision value measuring the likelihood that the input video belongs to the corresponding action class. Moreover, human actions are often associated with some specific natural environments and also exhibit high correlation with particular scene classes. It is therefore beneficial to utilize the contextual scene information for action recognition. In this paper, we build the high-level co-occurrence relationship between action classes and scene classes to discover the mutual contextual constraints between action and scene. By treating the scene class label as a latent variable, we propose to use the latent structural SVM (LSSVM) model to jointly capture the compatibility between multilevel action features (e.g., low-level visual context distribution feature and the corresponding mid-level class correlation feature) and action classes, the compatibility between multilevel scene features (i.e., SIFT feature and the corresponding class correlation feature) and scene classes, and the contextual relationship between action classes and scene classes. Extensive experiments on UCF Sports, YouTube and UCF50 datasets demonstrate the effectiveness of the proposed multilevel features and action-scene interaction based LSSVM model for human action recognition. Moreover, our method generally achieves higher recognition accuracy than other state-of-the-art methods on these datasets.

Object detection using Gaussian mixture-based optical flow modelling

This paper develops two background/foreground segmentation approaches based on a foreground subtraction from a background model, which uses scene colour and motion information. In the first approach, the background is modelled by a spatially global Gaussian mixture model based on scene red, green and blue colours. This model is then used to estimate motion-based optical flow, which helps indirectly in the scene segmentation decision. In an alternative approach, motion-based optical flow information is combined with colours as an augmented feature vector to model the background. For both approaches, we introduce an estimation method of the optical flow uncertainty statistics to use them in the background modelling. Evaluation results for both approaches based on indoor and outdoor image sequences show that the estimated background model is good at describing optical flow uncertainties and the segmentation obtained is better than colour only-based segmentation.

Human Gait Recognition Using Patch Distribution Feature and Locality-Constrained Group Sparse Representation

In this paper, we propose a new patch distribution feature (PDF) (i.e., referred to as Gabor-PDF) for human gait recognition. We represent each gait energy image (GEI) as a set of local augmented Gabor features, which concatenate the Gabor features extracted from different scales and different orientations together with the X-Y coordinates. We learn a global Gaussian mixture model (GMM) (i.e., referred to as the universal background model) with the local augmented Gabor features from all the gallery GEIs; then, each gallery or probe GEI is further expressed as the normalized parameters of an image-specific GMM adapted from the global GMM. Observing that one video is naturally represented as a group of GEIs, we also propose a new classification method called locality-constrained group sparse representation (LGSR) to classify each probe video by minimizing the weighted l(1, 2) mixed-norm-regularized reconstruction error with respect to the gallery videos. In contrast to the standard group sparse representation method that is a special case of LGSR, the group sparsity and local smooth sparsity constraints are both enforced in LGSR. Our comprehensive experiments on the benchmark USF HumanID database demonstrate the effectiveness of the newly proposed feature Gabor-PDF and the new classification method LGSR for human gait recognition. Moreover, LGSR using the new feature Gabor-PDF achieves the best average Rank-1 and Rank-5 recognition rates on this database among all gait recognition algorithms proposed to date.

Novel Gaussianized vector representation for improved natural scene categorization

We present a novel Gaussianized vector representation for scene images by an unsupervised approach. Each image is first encoded as an ensemble of orderless bag of features. A global Gaussian Mixture Model (GMM) learned from all images is then used to randomly distribute each feature into one Gaussian component by a multinomial trial. The posteriors of the feature on all the Gaussian components serve as the parameters of the multinomial distribution. Finally, the normalized means of the features distributed in every Gaussian component are concatenated to form a supervector, which is a compact representation for each scene image. We prove that these supervectors observe the standard normal distribution. The Gaussianized vector representation is a more generalized form of the widely used histogram representation. Our experiments on scene categorization tasks using this vector representation show significantly improved performance compared with the histogram-of-features representation. This paper is an extended version of our work that won the IBM Best Student Paper Award at the 2008 International Conference on Pattern Recognition (ICPR 2008) ( Zhou et al., 2008).

Modeling intrapersonal deformation subspace using GMM for palmprint identification

In this paper, an efficient model of palmprint identification is presented based on subspace density estimation using Gaussian Mixture Model (GMM). While a few training samples are available for each person, we use intrapersonal palmprint deformations to train the global GMM instead of modeling GMMs for every class. To reduce the dimension of such variations while preserving density function of sample space, Principle Component Analysis (PCA) is used to find the principle differences and form the Intrapersonal Deformation Subspace (IDS). After training GMM using Expectation Maximization (EM) algorithm in IDS, a maximum likelihood strategy is carried out to identify a person. Experimental results demonstrate the advantage of our method compared with traditional PCA method and single Gaussian strategy.

Segmental Approaches for Automatic Speaker Verification

Petrovska-Delacrétaz, Dijana, Černocký, Jan, Hennebert, Jean, and Chollet, Gérard, Segmental Approaches for Automatic Speaker Verification, Digital Signal Processing10(2000), 198–212.Speech is composed of different sounds (acoustic segments). Speakers differ in their pronunciation of these sounds. The segmental approaches described in this paper are meant to exploit these differences for speaker verification purposes. For such approaches, the speech is divided into different classes, and the speaker modeling is done for each class. The speech segmentation applied is based on automatic language independent speech processing tools that provide a segmentation of the speech requiring neither phonetic nor orthographic transcriptions of the speech data. Two different speaker modeling approaches, based on multilayer perceptrons (MLPs) and on Gaussian mixture models (GMMs), are studied. The MLP-based segmental systems have performance comparable to that of the global MLP-based systems, and in the mismatched train-test conditions slightly better results are obtained with the segmental MLP system. The segmental GMM systems gave poorer results than the equivalent global GMM systems.