Recent advances in rotation-invariant (RI) learning for 3D point clouds typically replace raw coordinates with handcrafted RI features to ensure robustness under arbitrary rotations. However, these approaches often suffer from the loss of global pose information, making them incapable of distinguishing geometrically similar but spatially distinct structures. We identify that this limitation stems from the restricted receptive field in existing RI methods, leading to Wing–tip feature collapse, a failure to differentiate symmetric components (e.g., left and right airplane wings) due to indistinguishable local geometries. To overcome this challenge, we introduce the Shadow-informed Pose Feature (SiPF), which augments local RI descriptors with a globally consistent reference point (referred to as the “shadow”) derived from a learned shared rotation. This mechanism enables the model to preserve global pose awareness while maintaining rotation invariance. We further propose Rotation-invariant Attention Convolution (RIAttnConv), an attention-based operator that integrates SiPFs into the feature aggregation process, thereby enhancing the model’s capacity to distinguish structurally similar components. Additionally, we design a task-adaptive shadow locating module based on the Bingham distribution over unit quaternions, which dynamically learns the optimal global rotation for constructing consistent shadows. Extensive experiments on 3D classification and part segmentation benchmarks demonstrate that our approach substantially outperforms existing RI methods, particularly in tasks requiring fine-grained spatial discrimination under arbitrary rotations.

With the widespread deployment of cross-modal retrieval in real-world scenarios, ensuring robustness against adversarial attacks is increasingly critical. Remarkably, deep cross-modal hashing is highly vulnerable to adversarial attacks due to its discrete nature and low-dimensional hash codes, while existing defense methods often fail to suppress perturbations embedded in vulnerable features and lack the capacity to model modality-specific structural differences, resulting in suboptimal adversarial robustness. To address these challenges, we propose a novel Disentangled Representation-Focused Generative Defense (DRFGD) framework for attack-tolerant cross-modal hashing. Without altering the structure of retrieval model, DRFGD defends against adversarial attacks by disentangling input representations into adversarial-robust and adversarial-vulnerable components, by an efficient dual-branch semantic-aware encoder. Guided by such disentangled robust features, an attack-tolerant generative module is seamlessly designed to synthesize semantically aligned and perturbation-resilient examples for robust adversarial training, thereby significantly promoting collaborative defense robustness to attackers. Consequently, the semantically consistent hash codes can be well obtained to enhance adversarial robustness in complex cross-modal attacking scenarios. Extensive experiments on public benchmarks demonstrate that DRFGD substantially improves retrieval robustness under various attacking scenarios, and shows its improved defense performance in comparison with the SOTA works.

Topic modeling plays a critical role in organizing and understanding large-scale web content. While neural topic models (NTMs) based on variational autoencoders (VAEs) have achieved notable success in analyzing textual data, they remain limited in addressing the multimodal nature of modern web content. Existing unimodal or multimodal extensions often suffer from posterior collapse and fail to capture the directional semantics inherent in both text and images, resulting in incoherent topics and limited interpretability. To address these challenges, we propose MM-vNTM (MultiModal Neural Topic Model with von Mises-Fisher Mixtures), a framework for web-scale topic discovery over multimodal data. MM-vNTM leverages pre-aligned cross-modal embeddings as inputs and jointly models document-level representations of text and image modalities in a shared hyperspherical latent space. Furthermore, it defines topics as mixtures of von Mises-Fisher (vMF) distributions in the L2-normalized word embedding space, explicitly capturing directional similarity. Experiments on multimedia web datasets demonstrate that MM-vNTM consistently outperforms state-of-the-art unimodal and multimodal baselines in terms of overall topic quality, highlighting its effectiveness for real-world web scenarios.

ABSTRACT Objectives In rapidly evolving industrial environments, resource management in Mobile Edge Computing (MEC) has gained increasing attention, aiming to ensure Quality of Service (QoS) for Artificial Intelligence of Things (AIoT) applications. While MEC reduces end‐to‐end delay, tasks offloaded to the cloud still encounter bottlenecks when processing massive AIoT‐generated data streams. To overcome this, we introduce a Collaborative Edge‐Edge (CE2) architecture that integrates heterogeneous edge servers and devices, enabling real‐time latency‐energy trade‐offs and accelerating AI‐driven decision‐making at the network edge. Managing resources in such dynamic, multi‐task, multi‐server environments remains challenging, especially under variable task‐arrival rates. Methods To tackle this, we propose LyDRM, a hybrid dynamic resource management scheme that synergistically combines model‐based optimization with model‐free deep reinforcement learning (DRL). A Lyapunov optimization module is embedded to enforce queue‐stability constraints, ensuring bounded task backlogs over time. Result To validate its effectiveness, extensive simulations show that LyDRM reduces the average weighted system cost‐defined as a combination of latency and energy metrics‐by at least 39.89%, significantly lowers both latency and energy consumption, accelerates convergence, and maintains long‐term stability in dynamic AIoT scenarios.

Spatial transcriptomics (ST) technologies have transformed our understanding of tissue biology by capturing gene expression with spatial context, enabling systematic analysis of cell-cell interactions (CCIs) and spatial domains in complex tissues. However, existing computational approaches often rely on fixed proximity graphs, curated ligand-receptor (LR) databases, or deep graph neural networks that are prone to over-smoothing and lack principled uncertainty quantification. These limitations hinder the discovery of heterogeneous, directional, and long-range CCIs essential for interpreting tissue organization and disease mechanisms. Here, we present B-HGME (Bayesian Hyperspherical Graph Mixture of Experts), a scalable, unsupervised framework that jointly delineates spatial domains and infers CCI networks from ST data with principled uncertainty estimates. B-HGME integrates spatial and gene-regulatory graphs into a dual-scale structure, encodes cell representations on a unit hypersphere via coupled message passing, and decodes edges using a Bayesian mixture-of-experts governed by a Dirichlet-regularized gating network. This design enables the model to capture multi-scale, directional, and biologically coherent interactions while avoiding the over-smoothing and posterior collapse of conventional models. The hyperspherical embedding geometry ensures angular similarity is preserved in high dimensions, and edge-level credibility is derived from the Bayesian posterior, facilitating interpretable and confident CCI inference. Across multiple datasets from six major ST platforms, B-HGME consistently achieves state-of-the-art spatial clustering accuracy and uncovers biologically coherent and diverse CCIs, including novel interactions beyond curated ligand-receptor pairs. B-HGME's hyperspherical embeddings accurately localize canonical astrocytic and laminar markers (e.g.,Gfap,Pcp4,Calb1, andCamk2a) to their expected spatial niches, confirming biochemical fidelity at single-gene resolution. Inferred ligand-receptor circuits not only recover known pathways but also reveal previously uncharacterized interactions (e.g., Astro1-eL2/3, VIP-Oligo), furnishing mechanistic hypotheses for cortical layer formation and tumor-stroma crosstalk. Together, these results demonstrate that B-HGME offers a powerful tool for spatial systems biology and hypothesis generation in development, immunity, and cancer. The source code of our model is available athttps://github.com/zxj8806/B-HGME.

In recent years, axial data, where observations are treated as axes of direction, has gained prominence in a range of complex tasks, including gene expression data clustering, blind speech separation, and depth image analysis. However, prevailing methods for axial data modeling mainly rely on shallow probabilistic models, which often overlook the hidden and hierarchical dependencies in the latent space. These methods also require a separate, human-engineered feature extractor to obtain features from raw axial data for downstream tasks. This work introduces a novel framework, Axial Variational Autoencoders (AVAEs), for modeling and representation learning of axial data by leveraging a deep generative model, the Variational Autoencoder (VAE). Unlike existing approaches, our method can autonomously learn more expressive representations from axial data by designing a VAE that uses the Watson distribution as the latent prior. Furthermore, we introduce a tailored reparameterization technique to support stable training. We validate the effectiveness of our model through experiments on simulated axial datasets and a real-world application.

The rapid integration of Artificial Intelligence (AI) and Internet of Things (IoT) technologies has led to the pro liferation of AIoT applications, significantly escalating demands for computing and communication resources in multi-user, multitask scenarios. A critical challenge lies in efficiently managing resource allocation to ensure Quality of Service (QoS) for diverse Deep Neural Network (DNN) inference tasks. Existing edge cloud collaborative inference approaches partially address this by hierarchical resource management; however, these methods often overlook the joint optimization of computation, communication, and data quality, and neglect long-term system stability in dynamic environments. To address these limitations, we propose EdgeManager, an online adaptive resource management frame work for hierarchical DNN inference in collaborative heterogeneous edge environments. Specifically, we formulate a Mixed Integer Nonlinear Programming (MINLP) optimization problem aimed at balancing inference accuracy and latency. Leveraging Lyapunov optimization, we transform the complex, multi-stage dynamic optimization problem into manageable deterministic sub-problems for each time slot, ensuring long-term stability. Furthermore, we introduce HyDRL-MO, a hybrid approach integrating model-free Deep Reinforcement Learning (DRL) and model-based multi-decision optimization techniques to achieve efficient and stable resource allocation. Extensive experimental evaluations demonstrate that EdgeManager significantly improves system performance, achieving up to 42.08% enhancement in average system benefits compared to state-of-the-art solutions.

Person detection in videos is vital for area admission and public safety. Existing studies have made significant progress in improving the accuracy of this task on the cloud. Meanwhile, with people's increasing awareness of privacy protection, there is a surging demand for privacy not being transmitted and processed by the cloud. Thus, providing services on edges becomes a promising solution. The dilemma is that edges are typically resource-constrained and cannot support the deployment of large models. However, tiny models that fit resource-constrained edges generally have unsatisfactory performance in accuracy and efficiency. To this end, we propose a Logical Correction Enabled Collaborative Person Detection Inference (LC-CPDI) framework for resource-constrained edges. First, we formulate the problem studied with a delay minimization objective. Second, we design a logical correction scheme to perceive abnormal predictions and perform corrections to improve accuracy. Third, a hybrid position prediction algorithm is proposed to replace time-consuming inference for simple scenarios. Finally, we design a collaborative inference scheme that enables frame outsourcing to idle edges to reduce the inference delay. We implemented LC-CPDI on a testbed designed with commercial edges. The experiments on real-world datasets show the effectiveness of LC-CPDI with up to 41.8% delay reduction on average and near 2% recall improvement.

Link prediction (LP) is fundamental to graph-based applications, yet existing graph autoencoders (GAEs) and variational GAEs (VGAEs) often struggle with intrinsic graph properties, particularly the presence of negative eigenvalues in adjacency matrices, which limits their adaptability and predictive performance. To address this limitation, we propose Hyperspherical Kolmogorov-Arnold Networks for LP (HKANLP), a novel framework that combines multiple graph neural network (GNN)-based representation learning strategies with Kolmogorov-Arnold networks (KANs) in a hyperspherical embedding space. Specifically, our model leverages the von Mises-Fisher (vMF) distribution to impose geometric consistency in the latent space and employs KANs as universal function approximators to reconstruct adjacency matrices, thereby mitigating the impact of negative eigenvalues and enhancing spectral diversity. Extensive experiments on homophilous, heterophilous, and large-scale graph datasets demonstrate that HKANLP achieves superior LP performance and robustness compared to state-of-the-art baselines. Furthermore, visualization analyses illustrate the model's effectiveness in capturing complex structural patterns. The source code of our model is publicly available at https://github.com/zxj8806/HKANLP/.

Medical AI models have achieved remarkable progress in various tasks, including medical image classification and segmentation. However, a critical issue frequently overlooked in clinical applications is the fairness of these models across different subgroups. Existing data collection strategies typically prioritize balancing disease categories while neglecting the importance of subgroup balance based on factors such as age and gender. This imbalance leads to the model's inconsistent performance across subgroups, thereby limiting its clinical applicability. This issue is particularly pronounced in skin cancer diagnosis due to the entanglement and complexity of feature patterns in skin cancer images. In this article, we approach the problem from a new perspective: making the feature patterns extracted by the model more independent and enhancing the representational ability of subgroups. We propose a novel framework that enhances feature independence across channels and improves subgroup representation. Central to our design is the Feature Pattern Separating Module (FPSM), combined with dual subgroupspecific classifiers and an auxiliary subgroup-type predictor. A diverse objective optimization model was introduced to guide the joint optimization of classification accuracy and subgroup fairness. Additionally, we proposed a novel classification loss function to dynamically balance the loss between subgroups. Experiments on two public skin cancer datasets demonstrate that our method improves both overall performance and fairness across subgroups, outperforming existing methods and offering a promising solution for fairer medical AI.

Topic modeling plays a vital role in uncovering latent semantic structures from large-scale biomedical corpora. While classical probabilistic models such as Latent Dirichlet Allocation (LDA) have been widely used, they often struggle with scalability and capturing complex semantic relationships in domain-specific contexts. Neural topic models (NTMs) based on variational autoencoders (VAEs) provide a more flexible and scalable alternative. However, existing NTMs face two key challenges: the neglect of the underlying geometric structure of semantic spaces and the lack of domain-specific adaptation, both of which contribute to suboptimal topic coherence and weak document clustering. To address these limitations, we pro-pose BiovMNVTM, a geometry-aware variational topic modeling framework tailored for biomedical text analysis. BiovMNVTM leverages the von Mises-Fisher (vMF) mixture distribution to model directional relationships on the hypersphere and integrates contextualized biomedical embeddings from BioBERT to capture rich, domain-specific semantics. Comprehensive experiments on multiple biomedical datasets show that BiovMNVTM achieves superior performance in terms of topic quality and clustering ability, demonstrating the effectiveness of incorporating geomet-ric and domain-aware modeling in biomedical topic discovery.

Deep neural networks (DNNs) have demonstrated exceptional performance across a variety of applications, yet they require substantial computing and power resources. In contrast, Spiking Neural Networks (SNNs) offer significant potential for energy-efficient computing due to their binary, event-driven properties. However, existing deep generative SNNs often struggle to produce high-quality low-dimensional representations in latent space, adversely affecting the quality of their generated samples. To address this limitation, we introduce a novel generative model that integrates the principles of variational autoencoders (VAEs) with adversarial training techniques. Our model consists of a generative module based entirely on an SNN-structured VAE and a discriminator employing artificial neural networks. This discriminator enhances the training process of our SNN-based generative model by applying adversarial principles similar to those used in Generative Adversarial Networks (GANs). Additionally, we have developed a conditional generation capability within our model, enabling the controlled production of specific images based on label inputs. Experimental evaluations on multiple datasets demonstrate that our model achieves superior image quality compared to existing top-performing SNN-based generative models. The source code of our model is accessible at https://github.com/zxj8806/SGM-VaGAN.

In this work, we propose a novel Generalized Zero-Shot Learning (GZSL) approach that integrates out-of-distribution (OOD) detection mechanisms into the architecture of generative models. At the core of our method is a dual-stream Variational Autoencoder (VAE), which creates a unified latent space to align data distributions from different modalities. This alignment helps our model better identify unseen classes by reducing bias using class-specific distribution ranges. We further improve the model by adding contrastive learning and a classification loss to the latent space, which separates features of seen and unseen classes and reduces feature confusion. To enhance the quality of generated data, we combine the dual-stream VAE with a Generative Adversarial Network (GAN), enhancing the quality of synthetic data and improving the model’s robustness against variations in data distribution. Extensive validation on popular benchmarks has confirmed the effectiveness of our approach, demonstrating notable improvements in GZSL performance.

The Audio-Visual Segmentation (AVS) task aims to identify sound-producing objects in the visual domain using auditory cues. Enhancing segmentation efficiency by incorporating prior knowledge, such as object locations and textual prompts, has proven to be crucial. However, existing methods suffer from feature misalignment during model training, leading to ineffective integration and reduced performance. To address this, we propose Triple-modalities alignment (TM-align), which combines audio signals, visual images, and textual prompts. By leveraging prompts from a frozen multi-modal large language model (MLLM), we extract two types of semantic information: contextual semantic description (C.S.D) and prompt specific summary (P.S.S). TM-align yields three pairs of aligned features: visual and C.S.D, visual and P.S.S, visual and audio, within two of our proposed cross-modalities alignment (CMA) models. To further enhance the alignment, we employ Jensen-Shannon Divergence (JSD) to regulate the domain distribution of the latter two features. By effectively aligning the three modalities, TM-align reduces redundancy and improves the overall AVS performance. Experimental results demonstrate that TM-align outperforms the mainstream AVS models.1

Person re-identification (ReID) in distributed surveillance networks presents significant networking challenges, particularly in coordinating multiple edge devices for real-time processing. While cloud-based solutions offer powerful computational capabilities, they introduce substantial network latency, hindering real-time performance in multi-camera indoor environments. To address these challenges, we present a distributed edge computing architecture that enables real-time person reidentification. Our system introduces two key networking innovations: (1) an$N$-frame feature matching mechanism that enhances identification accuracy at network edges, and (2) a dynamic load balancing framework that efficiently distributes processing tasks across edge nodes when network congestion occurs. As a foundation for this research, we introduce BNBUMTMC, a mixed dataset comprising both images and videos from multiple cameras, specifically tailored for indoor MTMCReID scenarios. Through extensive deployment in a university building environment with 10 cameras and multiple edge devices, we demonstrate that our system achieves high identification accuracy while significantly reducing network transmission latency compared to cloud-based approaches. Our experience provides practical insights into designing and implementing distributed edge computing systems for real-time surveillance applications.

Clustering high-dimensional directional data (i.e., L2 normalized vectors) presents significant challenges due to the intricate spherical representations of latent embeddings and the limitations of classical (non-deep) clustering techniques. Moreover, dynamically inferring the number of clusters remains a fundamental issue in existing deep clustering methods, especially those involving complex model-selection criteria. This paper addresses these challenges by introducing a novel deep nonparametric clustering framework that employs hyperspherical latent embeddings within a Variational Autoencoder architecture, enhanced by an infinite Von Mises-Fisher Mixture Model as a dynamic prior. This approach enables automatic adaptation of cluster numbers during training, eliminating the need for predefined clusters and traditional model selection processes. Our scalable architecture effectively integrates In-vMFMM with hyperspherical embeddings to tackle the complexities of directional data. Utilizing a joint training strategy, our method alternates between updating neural network parameters and adjusting mixture model priors via nonparametric variational Bayes. Empirical evaluations on benchmark datasets, including complex ImageNet-50, demonstrate that our approach significantly outperforms state-of-the-art deep nonparametric clustering methods. It also robustly estimates the number of clusters, showcasing its effectiveness and versatility in handling high-dimensional directional data.

The integration of edge computing with deep neural networks (DNNs) is crucial for intelligent industrial cyber-physical systems. Typically, deploying DNNs on heterogeneous edge devices relies on methods like model compression and partitioning. However, these approaches often result in homogeneous models across devices. This homogeneity limits the collective capability of edge computing systems, particularly in terms of generalization to diverse data distributions and adaptation to dynamic industrial environments. In this work, we propose to treat each DNN on an edge device as an independent model, aggregating their capabilities via ensemble learning to enhance generalization and dynamic adaptability. To realize this, we introduce the Adaptive Ensemble Knowledge Distillation Framework (AEKDF), combining cloud-based model training with edge computing based collaborative inference. In the cloud, AEKDF develops an enhanced Born Again Network that generates diverse, lightweight models tailored to specific edge devices through knowledge distillation. This process ensures model diversity which is critical to effective ensemble learning. On the edge, AEKDF employs an adaptive ensemble technique that aggregates prediction logits across devices, enabling rapid adaptation to changing environments and maintaining inference efficiency. Our extensive evaluations conducted on a realistic prototype demonstrate the substantial boost in predictive performance achieved by our AEKDF, showcasing a 4% to 10% accuracy improvement on the CIFAR-100 compared to conventional single-model approaches, while maintaining low latency.

This article proposes a novel deep clustering model based on the variational autoencoder (VAE), named GamMM-VAE, which can learn latent representations of training data for clustering in an unsupervised manner. Most existing VAE-based deep clustering methods use the Gaussian mixture model (GMM) as a prior on the latent space. We employ a more flexible asymmetric Gamma mixture model to achieve higher quality embeddings of the data latent space. Second, since the Gamma is defined for strictly positive variables, in order to exploit the reparameterization trick of VAE, we propose a transformation method from Gaussian distribution to Gamma distribution. This method can also be considered a Gamma distribution reparameterization trick, allows gradients to be backpropagated through the sampling process in the VAE. Finally, we derive the evidence lower bound (ELBO) based on the Gamma mixture model in an effective way for the stochastic gradient variational Bayesian (SGVB) estimator to optimize the proposed model. ELBO, a variational inference objective, ensures the maximization of the approximation of the posterior distribution, while SGVB is a method used to perform efficient inference and learning in VAEs. We validate the effectiveness of our model through quantitative comparisons with other state-of-the-art deep clustering models on six benchmark datasets. Moreover, due to the generative nature of VAEs, the proposed model can generate highly realistic samples of specific classes without supervised information.

In the fast-developing industrial environments, extensive focus on resource management within Mobile Edge Computing (MEC) aims to ensure low-latency QoS, however, some tasks offloaded to the cloud still experience high latency. Additionally, high energy consumption, poor link reliability, and excessive processing delays are intolerable for industrial applications. Compared to general servers, edge computing devices based on Arm architecture exhibit lower latency and higher energy efficiency. This highlights the need for improved heterogeneous Collaborative Edge-Edge Industrial Environments (CEIE) and precise multi-user QoS metrics. Thus, we focus on dynamic resource management within the CEIE architecture to better satisfy diverse industrial applications, formulating a multi-stage Mixed Integer Nonlinear Programming (MINLP) problem to minimize system costs. To reduce the computational complexity of solving the MINLP, we decompose the original problem into multi-user task offloading, Communication Resource Allocation (CmRA), and Computational Resource Allocation (CpRA) problems. These transformed problems are then tackled using DRMQ: an integrated learning optimization approach that combines model-free, priority experience replay-based Double Deep Q-Network (iDDQN) with model-based optimization, accelerating the Q-value function's convergence speed and reducing training time. Extensive simulations show that our proposed optimization scheme can reduce the average weighted system cost by at least 43.168% . Moreover, testbed experiments demonstrate that the proposed algorithm can reduce the average system cost by at least 42.650% in real-world applications, outperforming existing methods.

Semi-supervised learning (SSL) has been widely used to learn from both a few labeled images and many unlabeled images to overcome the scarcity of labeled samples in medical image segmentation. Most current SSL-based segmentation methods use pixel values directly to identify similar features in labeled and unlabeled data. They usually fail to accurately capture the intricate attachment structures in the left atrium, such as the areas of inconsistent density or exhibit outward curvatures, adding to the complexity of the task. In this paper, we delve into this issue and introduce an effective solution, CORAL(Correlation-Aligned)-Correlation Consistency Network (CORN), to capture the global structure shape and local details of Left Atrium. Diverging from previous methods focused on each local pixel value, the CORAL-Correlation Consistency Module (CCM) in the CORN leverages second-order statistical information to capture global structural features by minimizing the distribution discrepancy between labeled and unlabeled samples in feature space. Yet, direct construction of features from unlabeled data frequently results in "Sample Selection Bias", leading to flawed supervision. We thus further propose the Dynamic Feature Pool (DFP) for the CCM, which utilizes a confidence-based filtering strategy to remove incorrectly selected features and regularize both teacher and student models by constraining the similarity matrix to be consistent. Extensive experiments on the Left Atrium dataset have shown that the proposed CORN outperforms previous state-of-the-art semi-supervised learning methods.

normalized vectors) modeling has become a promising research topic in various real-world applications (such as gene expression data analysis, document categorization, and gesture recognition). In this work, we propose a hierarchical nonparametric Bayesian model based on von Mises-Fisher (VMF) distributions for modeling spherical data that involve multiple groups, where each observation within a group is sampled from a VMF mixture model with an infinite number of components allowing them to be shared across groups. Our model is formulated by employing a hierarchical nonparametric Bayesian framework known as the hierarchical Pitman-Yor (HPY) process mixture model, which possesses a power-law nature over the distribution of the components and is particularly useful for data distributions with heavy tails and skewness. To learn the proposed HPY process mixture model with VMF distributions, we systematically develop a closed-form optimization algorithm based on variational Bayes (VB). The merits of the proposed hierarchical Bayesian nonparametric model for modeling grouped spherical data are demonstrated through experiments on both synthetic data and a real-world application about resting-state functional magnetic resonance imaging (fMRI) data analysis.

In this article, we propose a predictive model based on Beta-Liouville (BL) and inverted BL (IBL) mixture models for occupancy estimation in smart buildings. This model gives better results than point estimate methods, when the training data is small, because it is based on data-driven predictive distribution. The Liouville-based mixture models were chosen because of their flexibility in fitting symmetric and asymmetric distributions. However, the large number of parameters of BL and IBL increases the uncertainty in approximating the upper bound of the predictive distribution, hence we propose an optimization scheme in which reliability is investigated and verified. In addition, we extend our work presented by giving more details about the predictive model and by studying the occupancy estimation in smart buildings problems in depth. Indeed, different occupancy scenarios are considered to show the merits of our predictive framework. This article aims to address the problem of occupancy estimation with a focus on scenarios where small training data sets are available. By developing robust predictive models that can generalize well with limited data, this research seeks to facilitate the early adoption and practical application of occupancy models in various domains.

Bayesian non-negative matrix factorization (BNMF) has been widely used in different applications. In this article, we propose a novel BNMF technique dedicated to semibounded data where each entry of the observed matrix is supposed to follow an Inverted Beta distribution. The model has two parameter matrices with the same size as the observation matrix which we factorize into a product of excitation and basis matrices. Entries of the corresponding basis and excitation matrices follow a Gamma prior. To estimate the parameters of the model, variational Bayesian inference is used. A lower bound approximation for the objective function is used to find an analytically tractable solution for the model. An online extension of the algorithm is also proposed for more scalability and to adapt to streaming data. The model is evaluated on five different applications: part-based decomposition, collaborative filtering, market basket analysis, transactions prediction and items classification, topic mining, and graph embedding on biomedical networks.

In recent years, clustering methods based on deep generative models have received great attention in various unsupervised applications, due to their capabilities for learning promising latent embeddings from original data. This article proposes a novel clustering method based on variational autoencoder (VAE) with spherical latent embeddings. The merits of our clustering method can be summarized as follows. First, instead of considering the Gaussian mixture model (GMM) as the prior over latent space as in a variety of existing VAE-based deep clustering methods, the von Mises-Fisher mixture model prior is deployed in our method, leading to spherical latent embeddings that can explicitly control the balance between the capacity of decoder and the utilization of latent embedding in a principled way. Second, a dual VAE structure is leveraged to impose the reconstruction constraint for the latent embedding and its corresponding noise counterpart, which embeds the input data into a hyperspherical latent space for clustering. Third, an augmented loss function is proposed to enhance the robustness of our model, which results in a self-supervised manner through the mutual guidance between the original data and the augmented ones. The effectiveness of the proposed deep generative clustering method is validated through comparisons with state-of-the-art deep clustering methods on benchmark datasets. The source code of the proposed model is available at https://github.com/fwt-team/DSVAE.

Abstract In this article, first, we propose a novel unsupervised learning method based on a hierarchical Dirichlet process mixture of shifted‐scaled Dirichlet (SSD) distributions. Second, we extend it to a hierarchical Pitman–Yor process mixture of SSD distributions. The goal is to find a model that properly fits complex real‐world data. Our models are based on SSD distributions that are more flexible than Dirichlet distribution in fitting proportional data. Simultaneous data fitting (parameter estimate) and model selection (model complexity determination) are possible with the suggested methods. We applied batch and online variational inference for learning the models. The online setting allows us to feed our models with large‐scale streaming data. The effectiveness of our proposed models is evaluated by four realistic and challenging applications, namely, spam email detection, texture clustering, traffic sign detection, and vehicle detection. Experimental results demonstrate the potential of our models to fit proportional data.

The target of generalized zero-shot learning (GZSL) is to train a model that can classify data samples from both seen categories and unseen categories under the circumstances that only the labeled samples from seen categories are available. In this paper, we propose a GZSL approach based on conditional generative models that adopts a contrastive disentanglement learning framework to disentangle visual information in the latent space. Specifically, our model encodes original and generated visual features into a latent space in which these visual features are disentangled into semantic-related and semantic-unrelated representations. The proposed contrastive learning framework leverages class-level and instance-level supervision, where it not only formulates contrastive loss based on semantic-related information at the instance level, but also exploits semantic-unrelated representations and the corresponding semantic information to form negative sample pairs at the class level to further facilitate disentanglement. Then, GZSL classification is performed by training a supervised model (e.g, softmax classifier) based only on semantic-related representations. The experimental results show that our model achieves state-of-the-art performance on several benchmark datasets, especially for unseen categories. The source code of the proposed model is available at: https://github.com/fwt-team/GZSL.

This paper aims at proposing an unsupervised hierarchical nonparametric Bayesian framework for modeling axial data (i.e., observations are axes of direction) that can be partitioned into multiple groups, where each observation within a group is sampled from a mixture of Watson distributions with an infinite number of components that are allowed to be shared across different groups. First, we propose a hierarchical nonparametric Bayesian model for modeling grouped axial data based on the hierarchical Pitman-Yor process mixture model of Watson distributions. Then, we demonstrate that by setting the discount parameters of the proposed model to 0, another hierarchical nonparametric Bayesian model based on hierarchical Dirichlet process can be derived for modeling axial data. To learn the proposed models, we systematically develop a closed-form optimization algorithm based on the collapsed variational Bayes (CVB) inference. Furthermore, to ensure the convergence of the proposed learning algorithm, an annealing mechanism is introduced to the framework of CVB inference, leading to an averaged collapsed variational Bayes inference strategy. The merits of the proposed models for modeling grouped axial data are demonstrated through experiments on both synthetic data and real-world applications involving gene expression data clustering and depth image analysis.

Clustering is a fundamental problem that frequently arises in many fields, such as pattern recognition, data mining, and machine learning. Although various clustering algorithms have been developed in the past, traditional clustering algorithms with shallow structures cannot excavate the interdependence of complex data features in latent space. Recently, deep generative models, such as autoencoder (AE), variational AE (VAE), and generative adversarial network (GAN), have achieved remarkable success in many unsupervised applications thanks to their capabilities for learning promising latent representations from original data. In this work, first we propose a novel clustering approach based on both Wasserstein GAN with gradient penalty (WGAN-GP) and VAE with a Gaussian mixture prior. By combining the WGAN-GP with VAE, the generator of WGAN-GP is formulated by drawing samples from the probabilistic decoder of VAE. Moreover, to provide more robust clustering and generation performance when outliers are encountered in data, a variant of the proposed deep generative model is developed based on a Student's-t mixture prior. The effectiveness of our deep generative models is validated though experiments on both clustering analysis and samples generation. Through the comparison with other state-of-art clustering approaches based on deep generative models, the proposed approach can provide more stable training of the model, improve the accuracy of clustering, and generate realistic samples.

Existing cross-modal hashing still faces three challenges: (1) Most batch-based methods are unsuitable for processing large-scale and streaming data. (2) Current online methods often suffer from insufficient semantic association, while lacking flexibility to learn the hash functions for varying streaming data. (3) Existing supervised methods always require much computation time or accumulate large quantization loss to learn hash codes. To address above challenges, we present an efficient Online Label Consistent Hashing (OLCH) for cross-modal retrieval, which aims to incrementally learn hash codes for the current arriving data, while updating the hash functions at a streaming manner. To be specific, an on-line semantic representation learning framework is designed to adaptively preserve the semantic similarity across different modalities, and a mini-batch online gradient descent approach associated with forward-backward splitting is developed to optimize the hash functions. Accordingly, the hash codes are adaptively learned online with the high discriminative capability, while avoiding high computation complexity to process the streaming data. Experimental results show its outstanding performance in comparison with the-state-of-arts.

In this paper, we propose hierarchical Pitman-Yor process mixtures of multivariate Beta distributions and learn this novel clustering method by online variational inference. The flexibility of this mixture model and its non-parametric hierarchical structure help in fitting our data. Also, the model complexity and model parameters are estimated simultaneously. We apply our proposed model to real medical applications. Our motivation is that labelling healthcare data is sensitive and expensive. Also, interpretability and evidence-based decision-making are some basic needs of medicine. These conditions led us to focus on clustering as it doesn’t need labelling. Another driving reason is that the amount of publicly available data in medicine is less compared to other fields due to the confidential regulations. To evaluate our proposed model, we compare its performance with other similar alternatives. The experimental results indicate the potential of our proposed model.

In this paper, we propose a clustering-based online news topic detection and tracking (TDT) approach based on hierarchical Bayesian nonparametric framework that allows topics to be shared across different news stories in a corpus. Our approach is formulated using the hierarchical Pitman-Yor process mixture model with the inverted Beta-Liouville (IBL) distribution as its component density, which has shown superior performance in modeling text data than the widely used Gaussian distribution. Moreover, we theoretically develop a convergence-guaranteed online learning algorithm that can effectively learn the proposed TDT model from a stream of news stories based on varational Bayes. The merits of our TDT approach are illustrated by comparing it with other well-defined clustering-based TDT approaches on different news data sets.

Fine-grained Image-text retrieval is challenging but vital technology in the field of multimedia analysis. Existing methods mainly focus on learning the common embedding space of images (or patches) and sentences (or words), whereby their mapping features in such embedding space can be directly measured. Nevertheless, most existing image-text retrieval works rarely consider the shared semantic concepts that potentially correlated the heterogeneous modalities, which can enhance the discriminative power of learning such embedding space. Toward this end, we propose a Cross-Graph Attention model (CGAM) to explicitly learn the shared semantic concepts, which can be well utilized to guide the feature learning process of each modality and promote the common embedding learning. More specifically, we build semantic-embedded graph for each modality, and smooth the discrepancy between two modalities via cross-graph attention model to obtain shared semantic-enhanced features. Meanwhile, we reconstruct image and text features via the shared semantic concepts and original embedding representations, and leverage multi-head mechanism for similarity calculation. Accordingly, the semantic-enhanced cross-modal embedding between image and text is discriminatively obtained to benefit the fine-grained retrieval with high retrieval performance. Extensive experiments evaluated on benchmark datasets show the performance improvements in comparison with state-of-the-arts.

In modern clinical medicine, electrocardiogram (ECG) is a common diagnosis technique of cardiovascular diseases. The purpose of this paper is to propose a novel model-based clustering approach for analyzing ECG data. Our approach is composed of two modules: representation learning and ECG data clustering. In the module of representation learning, a deep generative model referred to as the hyperspherical variational recurrent autoencoder (HVRAE) is developed to extract the representation of observed ECG data, based on the variational autoencoder (VAE) with long short-term memory (LSTM) networks. In the module of ECG data clustering, we develop a nonparametric hidden Markov model (NHMM) based on Dirichlet process in which the number of hidden states is inferred automatically during the learning process. Moreover, the emission density of each hidden state of our NHMM follows a mixture of von Mises-Fisher (VMF) distributions which have better capability for modeling ECG representations than other commonly used distributions (such as the Gaussian distribution). To learn the proposed VMF-based NHMM, we theoretically develop an effective learning algorithm based on variational Bayes. The merits of our model-based clustering approach for analyzing ECG data are verified through experiments on publicly available ECG data sets.

Many cognitive researches have shown the natural possibility of face-voice association, and such potential association has attracted much attention in biometric cross-modal retrieval domain. Nevertheless, the existing methods often fail to explicitly learn the common embeddings for challenging face-voice association tasks. In this paper, we present to learn discriminative joint embedding for face-voice association, which can seamlessly train the face subnetwork and voice subnetwork to learn their high-level semantic features, while correlating them to be compared directly and efficiently. Within the proposed approach, we introduce bi-directional ranking constraint, identity constraint and center constraint to learn the joint face-voice embedding, and adopt bi-directional training strategy to train the deep correlated face-voice model. Meanwhile, an online hard negative mining technique is utilized to discriminatively construct hard triplets in a mini-batch manner, featuring on speeding up the learning process. Accordingly, the proposed approach is adaptive to benefit various face-voice association tasks, including cross-modal verification, 1:2 matching, 1:N matching, and retrieval scenarios. Extensive experiments have shown its improved performances in comparison with the state-of-the-art ones.

Road extraction from remote sensing images is an attractive but difficult task. Gray-value distribution and structure feature information are both crucial for road extraction task. However, existing methods mainly focus on structure feature information which contains morphological shape features and machine learning features, suffering from lots of false positives which are generated at positions having similar structure features but different gray-value distribution with roads. To effectively fuse the two complementary gray-value distribution and structure feature information, we propose a coarse-to-fine road extraction algorithm from remote sensing images. First, at the coarse level, we introduce a local Dirichlet mixture models (LDMM) which utilizing gray-value distribution information to pre-segment images into potential roads and backgrounds. Thus, most backgrounds having different gray-value distribution with roads can be removed firstly. Compared with original Dirichlet mixture models, the LDMM is much faster and more accurate. Next, at the fine level, we introduce a multiscale-high-order deep learning strategy based on ResNet model which can learn robust structure context features for final road extraction step. Based on the results of LDMM, the multiscale-high-order strategy can further remove false positives which have different structure features with roads. Compared with a single scanning size ResNet, our multiscale-high-order strategy can learn higher-order context information, leading to better performances. We test our algorithm on Shaoshan dataset. Experiments illustrate our better performance compared with other six state-of-the-art methods.

In this article, we propose an effective mixture model-based approach to modeling and clustering positive data vectors. Our mixture model is based on the inverted Beta-Liouville (IBL) distribution which is extracted from the family of Liouville distributions. To cope with the problem of determining the appropriate number of clusters in our approach, a nonparametric Bayesian framework is used to extend the IBL mixture to an infinite mixture model in which the number of clusters is assumed to be infinite initially and will be inferred automatically during the learning process. To optimize the proposed model, we propose a convergence-guaranteed learning algorithm based on the averaged collapsed variational Bayes inference that can effectively learn model parameters with closed-form solutions. The effectiveness of the proposed infinite IBL mixture model for modeling and clustering positive vectors is validated through both synthetic and real-world data sets.

Cross-modal hashing has recently gained significant popularity to facilitate retrieval across different modalities. With limited label available, this paper presents a novel Semi-Supervised Semantic-Preserving Hashing (S3PH) for flexible cross-modal retrieval. In contrast to most semi-supervised cross-modal hashing works that need to predict the label of unlabeled data, our proposed approach groups the labeled and unlabeled data together, and integrates the relaxed latent subspace learning and semantic-preserving regularization across different modalities. Accordingly, an efficient relaxed objective function is proposed to learn the latent subspaces for both labeled and unlabeled data. Further, an orthogonal rotation matrix is efficiently learned to transform the latent subspace to hash space by minimizing the quantization error. Without sacrificing the retrieval performance, the proposed S3PH method can benefit various kinds of retrieval tasks, i.e., unsupervised, semi-supervised and supervised. Experimental results compared with several competitive algorithms show the effectiveness of the proposed method and its superiority over state-of-the-arts.

This paper proposes a Bayesian nonparametric framework for clustering axially symmetric data. Our approach is based on a Dirichlet processes mixture model with Watson distributions, which can also be considered as the infinite Watson mixture model. In this paper, first, we extend the finite Watson mixture model into its infinite counterpart based on the framework of truncated Dirichlet process mixture model with a stick-breaking representation. Second, we propose a coordinate ascent mean-field variational inference algorithm that can effectively learn the parameters of our model with closed-form solutions; Third, to cope with a massive data set, we develop a stochastic variational inference algorithm to learn the proposed model through the method of stochastic gradient ascent; Finally, the proposed nonparametric Bayesian model is evaluated through simulated axially symmetric data sets and a real-world application, namely, gene expression data clustering.

A novel framework is developed for the modelling and clustering of proportional data (i.e. normalised histograms) based on the Beta‐Liouville mixture model. This framework is based on incremental model selection, by testing if a given component was truly Beta‐Liouville distributed. Specifically, the authors compare the theoretical maximum entropy of the given component with the estimated entropy obtained by the MeanNN estimator. If a significant difference was gained from this comparison, this component is considered as not well fitted and is then splitted into two new components with a proper initialisation. Our approach is tested through synthetic data sets and real‐world applications which involve human gesture recognition and vehicle tracking for traffic monitoring purposes, which demonstrate that the authors' approach is superior to comparable techniques.

The availability of large motion capture (mocap) data has sparked a great motivation for computer animation, and the task of automatically annotating complex mocap sequences plays an important role in the efficient motion analysis. To this end, this study presents an efficient human mocap data annotation approach by using multi‐view spatiotemporal feature fusion. First, the authors exploit an improved hierarchical aligned cluster analysis algorithm to divide the unknown human mocap sequence into several sub‐motion clips, and each sub‐motion clip incorporates a particular semantic meaning. Then, the two kinds of multi‐view features, namely most informative central distances and most informative geometric angles, are discriminatively extracted and temporally modelled by a Fourier temporal pyramid to complementarily characterise each motion clip. Finally, the authors utilise the discriminant correlation analysis to fuse these two types of motion features and further employ an extreme learning machine to annotate each sub‐motion clip. The extensive experiments tested on the public available database have demonstrated the effectiveness of the proposed approach in comparison with the existing counterparts.

In this paper, a hierarchical Dirichlet process (HDP) mixture model of generalized inverted Dirichlet (GID) distributions with an unsupervised feature selection scheme is developed. The proposed model is learned via a principled variational framework and then deployed for video modeling and segmentation. Experimental results show the merits of our developed statistical framework.

In this paper, a novel statistical generative model based on hierarchical Pitman-Yor process and generalized Dirichlet distributions (GDs) is presented. The proposed model allows us to perform joint clustering and feature selection thanks to the interesting properties of the GD distribution. We develop an online variational inference algorithm, formulated in terms of the minimization of a Kullback-Leibler divergence, of our resulting model that tackles the problem of learning from high-dimensional examples. This variational Bayes formulation allows simultaneously estimating the parameters, determining the model's complexity, and selecting the appropriate relevant features for the clustering structure. Moreover, the proposed online learning algorithm allows data instances to be processed in a sequential manner, which is critical for large-scale and real-time applications. Experiments conducted using challenging applications, namely, scene recognition and video segmentation, where our approach is viewed as an unsupervised technique for visual learning in high-dimensional spaces, showed that the proposed approach is suitable and promising.

Beta-Liouville mixture models have achieved measurable success in many computer vision and pattern recognition applications. In this paper, we develop a novel algorithm to learn this particular kind of models that have been shown to be very efficient for the clustering of proportional data. Our algorithm is based on an accelerated version of the variational Bayes approach. Experiments show that the developed algorithm work very well for the categorization of 3D shapes.

In this paper, we formulate 3D model recognition as a statistical inference problem using a Pitman-Yor process mixture of Beta-Liouville Distributions. The proposed model is learned via a collapsed variational inference approach. Unlike classic variational Bayes, the collapsed approach does not make the non-realistic assumption that the model's parameters are independent from the assignment variables, which leads to better modelling and generalization capabilities. The merits and advantages of the proposed approach are shown via extensive experiments.

This paper proposes a hierarchical Pitman-Yor (HPY) process mixture of Dirichlet distributions. It can be viewed as an extension of our previous works that have considered Dirich-let process mixtures of Dirichlet distributions (i.e. infinite Dirichlet mixture models). The proposed model is learned via a variational Bayes approach and applied to the challenging problem of dynamic textures clustering.

We propose model-based inference for topic novelty detection using a non-parametric Bayesian probability model. The probability model is a Dirichlet process mixture of inverted Dirichlet distributions which can be viewed as an infinite mixture model. The inference is based on variational Bayes deployed using approximate conjugate priors to the inverted Dirichlet. Detailed experimental study demonstrates the merits of our approach and shows that it gives good description of the data.

Exploiting Dirichlet process mixture models (also known as infinite mixture models) to model visual and textual data is now standard weapon in the arsenal of machine learning. This paper proposes a new accelerated variational inference approach to learn Dirichlet process mixture models with Dirichlet distributions. The choice of using Dirichlet distribution as the basic distribution is mainly due to its flexibility for modeling proportional data. Indeed, this kind of data is naturally generated by several applications involving the representation of texts, images and videos using the bag-of-words (or "visual words" in the case of images and videos) approach. The potential of the developed learning framework is shown using a challenging real application namely human action recognition in videos.

Recently, finite Beta-Liouville mixture models have proved to be an effective and powerful knowledge representation and inference engine in several machine learning and data mining applications. In this paper, we propose a component splitting and local model selection method to address the problem of learning and selecting finite Beta-Liouville mixture models in an incremental variational way. Within the proposed principled variational learning framework, all the involved parameters and model complexity (i.e. the number of mixture components) can be estimated simultaneously in a closed-form. We demonstrate the effectiveness of the proposed approach through both synthetic data as well as two challenging real-world applications namely human activities modeling and recognition, and facial expressions recognition.

In this paper, an expectation propagation (EP) inference framework for unsupervised feature selection is proposed for modeling proportional data which naturally appear in many applications such as text and image modeling, in the context of finite mixture-based clustering. Within our framework, simultaneous clustering and feature selection is formalized using finite mixtures of generalizing Dirichlet (GD) distributions. The proposed EP-based inference framework allows to obtain a full posterior distribution on all our unsupervised feature selection model's parameters. Moreover, the complexity of the deployed mixture models and all the involved model parameters can be evaluated simultaneously. The effectiveness and efficiency of the proposed algorithm are evaluated on both synthetic data and two challenging applications namely human action videos categorization and facial expression recognition.

Clustering has been a subject of extensive research in data mining, pattern recognition, and other areas for several decades. The main goal is to assign samples, which are typically non-Gaussian and expressed as points in high-dimensional feature spaces, to one of a number of clusters. It is well known that in such high-dimensional settings, the existence of irrelevant features generally compromises modeling capabilities. In this paper, we propose a variational inference framework for unsupervised non-Gaussian feature selection, in the context of finite generalized Dirichlet (GD) mixture-based clustering. Under the proposed principled variational framework, we simultaneously estimate, in a closed form, all the involved parameters and determine the complexity (i.e., both model an feature selection) of the GD mixture. Extensive simulations using synthetic data along with an analysis of real-world data and human action videos demonstrate that our variational approach achieves better results than comparable techniques.

A large class of problems can be formulated in terms of the clustering process. Mixture models are an increasingly important tool in statistical pattern recognition and for analyzing and clustering complex data. Two challenging aspects that should be addressed when considering mixture models are how to choose between a set of plausible models and how to estimate the model's parameters. In this paper, we address both problems simultaneously within a unified online nonparametric Bayesian framework that we develop to learn a Dirichlet process mixture of Beta-Liouville distributions (i.e., an infinite Beta-Liouville mixture model). The proposed infinite model is used for the online modeling and clustering of proportional data for which the Beta-Liouville mixture has been shown to be effective. We propose a principled approach for approximating the intractable model's posterior distribution by a tractable one-which we develop-such that all the involved mixture's parameters can be estimated simultaneously and effectively in a closed form. This is done through variational inference that enjoys important advantages, such as handling of unobserved attributes and preventing under or overfitting; we explain that in detail. The effectiveness of the proposed work is evaluated on three challenging real applications, namely facial expression recognition, behavior modeling and recognition, and dynamic textures clustering.

Online algorithms allow data points to be processed sequentially, which is important for real-time applications. In this paper, we propose a novel online clustering approach based on a mixture of Dirichlet processes with Dirichlet distributions, which can be viewed as an extension of the finite Dirichlet mixture model to the infinite case. Our approach is based on nonparametric Bayesian analysis, the determination of the number of clusters is sidestepped by assuming an infinite number of clusters. By learning the proposed model in an online manner with a variational learning framework, all the involved parameters can be estimated effectively and efficiently in a closed form without introducing the problem of over fitting. The proposed online infinite mixture model is validated through both synthetic data sets and a challenging real-world application namely unsupervised image categorization.

In this paper, we focus on the variational learning of finite Dirichlet mixture models. Compared to other algorithms that are commonly used for mixture models (such as expectation-maximization), our approach has several advantages: first, the problem of over-fitting is prevented; furthermore, the complexity of the mixture model (i.e., the number of components) can be determined automatically and simultaneously with the parameters estimation as part of the Bayesian inference procedure; finally, since the whole inference process is analytically tractable with closed-form solutions, it may scale well to large applications. Both synthetic and real data, generated from real-life challenging applications namely image databases categorization and anomaly intrusion detection, are experimented to verify the effectiveness of the proposed approach.

In recent years, an increasing number of security threats have brought a serious risk to the internet and computer networks. Intrusion Detection System (IDS) plays a vital role in detecting various kinds of attacks. Developing adaptive and flexible oriented IDSs remains a challenging and demanding task due to the incessantly appearance of new types of attacks and sabotaging approaches. In this paper, we propose a novel unsupervised statistical approach for detecting network based attacks. In our approach, patterns of normal and intrusive activities are learned through finite generalized Dirichlet mixture models, in the context of Bayesian variational inference. Under the proposed variational framework, the parameters, the complexity of the mixture model, and the features saliency can be estimated simultaneously, in a closed-form. We evaluate the proposed approach using the popular KDD CUP 1999 data set. Experimental results show that this approach is able to detect many different types of intrusions accurately with a low false positive rate.

Video texture is a new type of medium which can provide a new video with a continuously varying stream of images from a recorded video. It is created by reordering the input video frames in a way which can be played without any visual discontinuity. Recently, a new method of generating video textures has been proposed. It first apply principal components analysis (PCA) to extract signatures or patterns from the original video sequence, and then implement an autoregressive process (AR) model to synthesize new video textures. In this paper, we extend this video texture generation method by comparing PCA with other dimensionality reduction techniques such as probabilistic principal components analysis, kernel principal components analysis, independent component analysis, local linear embedding and Isomap. According to our experiments, these approaches prevail the original approach by providing us video textures with better quality.