Sparse Reconstructive Evidential Clustering for Multi-View Data

Although many multi-view clustering(MVC) algorithms with acceptable performances have been presented, to the best of our knowledge, nearly all of them need to be fed with the correct number of clusters. In addition, these existing algorithms create only the hard and fuzzy partitions for multi-view objects,which are often located in highly-overlapping areas of multi-view feature space. The adoption of hard and fuzzy partition ignores the ambiguity and uncertainty in the assignment of objects, likely leading to performance degradation. To address these issues, we propose a novel sparse reconstructive multi-view evidential clustering algorithm(SRMVEC). Based on a sparse reconstructive procedure, SRMVEC learns a shared affinity matrix across views, and maps multi-view objects to a 2-dimensional humanreadable chart by calculating 2 newly defined mathematical metrics for each object. From this chart, users can detect the number of clusters and select several objects existing in the dataset as cluster centers. Then, SRMVEC derives a credal partition under the framework of evidence theory, improving the fault tolerance of clustering. Ablation studies show the benefits of adopting the sparse reconstructive procedure and evidence theory. Besides,SRMVEC delivers effectiveness on benchmark datasets by outperforming some state-of-the-art methods.

Contrastive Consistency and Attentive Complementarity for Deep Multi-View Subspace Clustering

Deep multi-view subspace clustering (DMVSC) based on self-expression has attracted increasing attention dueto its outstanding performance and nonlinear application. However, most existing methods neglect that viewprivatemeaningless information or noise may interfere with the learning of self-expression, which may lead to thedegeneration of clustering performance. In this paper, we propose a novel framework of Contrastive Consistencyand Attentive Complementarity (CCAC) for DMVsSC. CCAC aligns all the self-expressions of multiple viewsand fuses them based on their discrimination, so that it can effectively explore consistent and complementaryinformation for achieving precise clustering. Specifically, the view-specific self-expression is learned by a selfexpressionlayer embedded into the auto-encoder network for each view. To guarantee consistency across views andreduce the effect of view-private information or noise, we align all the view-specific self-expressions by contrastivelearning. The aligned self-expressions are assigned adaptive weights by channel attention mechanism according totheir discrimination. Then they are fused by convolution kernel to obtain consensus self-expression withmaximumcomplementarity ofmultiple views. Extensive experimental results on four benchmark datasets and one large-scaledataset of the CCAC method outperformother state-of-the-artmethods, demonstrating its clustering effectiveness.

Low-Rank Multi-View Subspace Clustering Based on Sparse Regularization

Multi-view Subspace Clustering (MVSC) emerges as an advanced clustering method, designed to integrate diverse views to uncover a common subspace, enhancing the accuracy and robustness of clustering results. The significance of low-rank prior in MVSC is emphasized, highlighting its role in capturing the global data structure across views for improved performance. However, it faces challenges with outlier sensitivity due to its reliance on the Frobenius norm for error measurement. Addressing this, our paper proposes a Low-Rank Multi-view Subspace Clustering Based on Sparse Regularization (LMVSC- Sparse) approach. Sparse regularization helps in selecting the most relevant features or views for clustering while ignoring irrelevant or noisy ones. This leads to a more efficient and effective representation of the data, improving the clustering accuracy and robustness, especially in the presence of outliers or noisy data. By incorporating sparse regularization, LMVSC-Sparse can effectively handle outlier sensitivity, which is a common challenge in traditional MVSC methods relying solely on low-rank priors. Then Alternating Direction Method of Multipliers (ADMM) algorithm is employed to solve the proposed optimization problems. Our comprehensive experiments demonstrate the efficiency and effectiveness of LMVSC-Sparse, offering a robust alternative to traditional MVSC methods.

Virtual Epipolar Line Construction of Single-Lens Bi-Prism Stereovision System

This paper proposes a simple geometrical ray based approach to solve the stereo correspondence problem for the single-lens bi-prism stereovision system. Each image captured using this system can be divided into two sub-images on the left and right and these sub-images are generated by two virtual cameras which are produced by the bi-prism. This stereovision system is equivalent to the conventional two camera system and the two sub-images captured have disparities which can be used to reconstruct back the 3-dimensional （3D） scene. The stereo correspondence problem of this system will be solved geometrically by applying the epipolar geometry constraint on the generated virtual cameras instead of the real CCD camera. Experiments are conducted to validate the proposed method and the results are compared to the calibration based approach to confirm its accuracy and effectiveness.

Automatic teaching of stereovision-guided welding robot using ant colony optimization algorithm

A binocular stereovision system with a linear laser emitter is developed to detect seam position and its orientation, employing acquired 3-dimensional seam data, the automatic teaching of welding robot is implemented using a controlling strategy based on ant colony optimization（ ACO ） algorithm, in which the angle increment of robot joint is discretized as the nodes of ACO graph and a corresponding pheromone updating strategy is presented. The experimental results for curvilinear seams and saddle-shaped seams show that the automatic teaching of welding robot can be successfully completed using the ACO-based controlling strategy under the guidance of stereovision, and the welding trajectory generated by the proposed method has higher accuracy and less setting time compared with conventional proportional-integral-differential （PID） controller and fuzzy controller.

A simple rectification method for linear multi-baseline stereovision system

The linear multi-baseline stereo system introduced by the CMU-RI group has been proven to be a very effective and robust stereovision system. However, most traditional stereo rectification algorithms are all designed for binocular stereovision system, and so, cannot be applied to a linear multi-baseline system. This paper presents a simple and intuitional method that can simultaneously rectify all the cameras in a linear multi-baseline system. Instead of using the general 8-parameter homography transform, a two-step virtual rotation method is applied for rectification, which results in a more specific transform that has only 3 parameters, and more stability. Experimental results for real stereo images showed the presented method is efficient.

AUTOMATED IMAGE MATCHING WITH CODED POINTS IN STEREOVISION MEASUREMENT

A coding-based method to solve the image matching problems in stereovision measurement is presented. The solution is to add and append an identity ID to the retro-reflect point, so it can be identified efficiently under the complicated circumstances and has the characteristics of rotation, zooming, and deformation independence. Its design architecture and implementation process in details based on the theory of stereovision measurement are described. The method is effective on reducing processing data time, improving accuracy of image matching and automation of measuring system through experiments.

Visualizing surface strain distribution of facial skin using stereovision

An experimental technique has been developed for measuring and visualizing strain distribution on facial skin. A stereovision technique based on digital image correlation is employed for obtaining the displacement distribution on the human face. Time-variation of the movement of the facial skin surface is obtained from consecutive images obtained using a pair of high-speed cameras. The strains on the facial skin surface are then obtained from the measured displacements. The performance of the developed system is demonstrated by applying it to the measurement of the strain on facial skin during the production of sound. Results show that the strains on facial skin can be visualized. Further discussion on the relationship between the creation of wrinkles and strains is possible with the help of the developed system.

Crowdedness estimation approach based on stereovision for bus passengers

An estimation approach is proposed in this paper based on the binocular stereovision to collect the degree of crowdedness in public transports. The proposed method combines the disparity with frame differences to extract the foreground object. An adaptive window normalized cross correlation （NCC） matching and interpolated method is applied to get the sub-pixel image disparity value. Then, the foreground object is projected to the horizontal plane to eliminate the influence of the occlusion and perspective effect. Finally the degree of crowdedness is calculated from the area and the perimeter of the foreground objects. Experimental results show that the proposed method can obtain good estimation results in the simulated scenes in the laboratory and on parking or moving buses. This approach is effective to illumination changes, shadows and occlusion of passengers.

AB019. A proposed test of stereovision to identify local retinal abnormalities

Background:The human visual system extracts depth information from disparity in the images seen by the two eyes.The ability to calculate depth from disparity will be disrupted if local retinal abnormalities distort parts of those images,especially if these distortions are different in the two eyes.In its early stages,age-related macular degeneration(AMD)causes slight distortions in the central vision field which differ in the two eyes.AMD is the most common form of irreversible blindness in people over the age of 50.The goal of this project is to develop a stereoscopic perception test which leverages the sensitivity of binocular depth perception to detect the interocular differences symptomatic of either early-stage AMD or other diseases affecting the retina.Methods:A program was written in MATLAB that allowed separate left and right eye stimuli to be shown to the two eyes.NVIDIA 3D Vision 2 stereoscopic glasses were used to present the stimuli.The test we have developed consists of random dot patterns covering the central 5 degrees of vision.One or more disk-shaped perturbations in depth are displayed at different locations in the visual field of the subjects.Of the ten possible target locations,we present between one and four disks on each trial.The disks will only be visible if there is an undistorted input for that visual field location from the two retinae.The participant uses a keypad to report the number of floating disks seen.A set of trials with randomized locations and numbers of disks is used to gather initial data on likely areas of stereoscopic vision deficit;afterwards,likelihoods for deficits in each location are calculated and used to generate customized subsequent trials.Results:The software to perform the local stereovision test has been developed and is now being piloted.We are currently collecting data from healthy normal subjects prior to applying the test to clinical populations.In order to simulate the central vision distortions of AMD,patches of the stimulus for one eye are scrambled and blurred.This allows us to ensure that the task is functioning correctly.Conclusions:The next step for this project is for it to be tested on ageing clinical populations to find its effectiveness in differentiating patients with normal sight from those showing early symptoms of AMD or other retinal abnormalities such as diabetic retinopathy.This test could represent a novel approach in early clinical detection of ocular disease.

Multi-View & Transfer Learning for Epilepsy Recognition Based on EEG Signals

Epilepsy is a central nervous system disorder in which brain activity becomes abnormal.Electroencephalogram(EEG)signals,as recordings of brain activity,have been widely used for epilepsy recognition.To study epilep-tic EEG signals and develop artificial intelligence(AI)-assist recognition,a multi-view transfer learning(MVTL-LSR)algorithm based on least squares regression is proposed in this study.Compared with most existing multi-view transfer learning algorithms,MVTL-LSR has two merits:(1)Since traditional transfer learning algorithms leverage knowledge from different sources,which poses a significant risk to data privacy.Therefore,we develop a knowledge transfer mechanism that can protect the security of source domain data while guaranteeing performance.(2)When utilizing multi-view data,we embed view weighting and manifold regularization into the transfer framework to measure the views’strengths and weaknesses and improve generalization ability.In the experimental studies,12 different simulated multi-view&transfer scenarios are constructed from epileptic EEG signals licensed and provided by the Uni-versity of Bonn,Germany.Extensive experimental results show that MVTL-LSR outperforms baselines.The source code will be available on https://github.com/didid5/MVTL-LSR.

ER-Net:Efficient Recalibration Network for Multi-ViewMulti-Person 3D Pose Estimation

Multi-view multi-person 3D human pose estimation is a hot topic in the field of human pose estimation due to its wide range of application scenarios.With the introduction of end-to-end direct regression methods,the field has entered a new stage of development.However,the regression results of joints that are more heavily influenced by external factors are not accurate enough even for the optimal method.In this paper,we propose an effective feature recalibration module based on the channel attention mechanism and a relative optimal calibration strategy,which is applied to themulti-viewmulti-person 3D human pose estimation task to achieve improved detection accuracy for joints that are more severely affected by external factors.Specifically,it achieves relative optimal weight adjustment of joint feature information through the recalibration module and strategy,which enables the model to learn the dependencies between joints and the dependencies between people and their corresponding joints.We call this method as the Efficient Recalibration Network(ER-Net).Finally,experiments were conducted on two benchmark datasets for this task,Campus and Shelf,in which the PCP reached 97.3% and 98.3%,respectively.

Relational graph location network for multi-view image localization

In multi-view image localization task,the features of the images captured from different views should be fused properly.This paper considers the classification-based image localization problem.We propose the relational graph location network(RGLN)to perform this task.In this network,we propose a heterogeneous graph construction approach for graph classification tasks,which aims to describe the location in a more appropriate way,thereby improving the expression ability of the location representation module.Experiments show that the expression ability of the proposed graph construction approach outperforms the compared methods by a large margin.In addition,the proposed localization method outperforms the compared localization methods by around 1.7%in terms of meter-level accuracy.

Diverse Deep Matrix Factorization With Hypergraph Regularization for Multi-View Data Representation

Deep matrix factorization(DMF)has been demonstrated to be a powerful tool to take in the complex hierarchical information of multi-view data(MDR).However,existing multiview DMF methods mainly explore the consistency of multi-view data,while neglecting the diversity among different views as well as the high-order relationships of data,resulting in the loss of valuable complementary information.In this paper,we design a hypergraph regularized diverse deep matrix factorization(HDDMF)model for multi-view data representation,to jointly utilize multi-view diversity and a high-order manifold in a multilayer factorization framework.A novel diversity enhancement term is designed to exploit the structural complementarity between different views of data.Hypergraph regularization is utilized to preserve the high-order geometry structure of data in each view.An efficient iterative optimization algorithm is developed to solve the proposed model with theoretical convergence analysis.Experimental results on five real-world data sets demonstrate that the proposed method significantly outperforms stateof-the-art multi-view learning approaches.

基于双目立体视觉的乘员运动姿态测量方法研究

提出了基于双目立体视觉的汽车乘员运动姿态测量方法。首先使用自主搭建的双目立体视觉系统,按照张正友标定法标定相机并对图像进行立体校正;然后提出基于邻域的SGBM单参数优化算法,并依此建立乘员关键特征点与三维空间坐标间的映射关系,实现了对乘员运动姿态的非接触动态测量。实验结果表明,该方法可以有效地测量乘员运动姿态,且测量精度较高,对汽车乘员非标准坐姿下的保护策略制定具有重要的参考价值。

基于关键点密度优化的ORB算法

在立体视觉检测系统中,特征匹配技术至关重要,其用于识别和对齐不同图像间的相似特征,实现图像对比、物体识别、三维重建等多项任务。特征匹配的质量直接影响整个立体视觉检测系统的精度,特征点提取是特征匹配的基础,这些点的质量直接决定了匹配的准确性和算法的鲁棒性。ORB算法因具有高效性,被广泛应用于特征匹配任务,但在处理复杂场景时,特征点在数量和分布均匀性方面存在不足。对此,提出了一种改进的基于关键点密度的自适应抽样方法,通过结合图像的局部对比度和梯度信息,优化ORB算法中关键点的分布,以实现整个图像中关键点的均匀选取,提高特征点提取性能。利用Middlebury立体视觉数据集进行的实验结果表明,改进后的算法相比传统方法,在关键点数量和分布均匀性上有显著提升,同时保持了接近原ORB算法的运行效率。此项研究不仅针对ORB算法在复杂场景处理中的不足提供了有效的解决方案,也为计算机视觉领域特征点提取和匹配的优化开辟了新的途径。

基于光栅投影双目立体视觉的综合实践设计

基于光栅投影双目立体视觉的综合实践设计旨在培养学生在仪器科学与技术领域的专业技能和实践能力。首先在课程教学中引入思政教育;其次引导学生搭建实验平台、完成相机标定,通过MATLAB编码求解包裹相位、相位解包裹、三维点云生成等,在教学过程中采用项目驱动教学,分模块进行实践教学。最后,通过课程实践培养学生的创新能力和解决复杂问题的能力。

Interactive transport of multi-view videos for 3DTV applications

The authors propose a novel method for transporting multi-view videos that aims to keep the bandwidth requirements on both end-users and servers as low as possible. The method is based on application layer multicast, where each end point re- ceives only a selected number of views required for rendering video from its current viewpoint at any given time. The set of selected videos changes in real time as the user’s viewpoint changes because of head or eye movements. Techniques for reducing the black-outs during fast viewpoint changes were investigated. The performance of the approach was studied through network experiments.

Multi-view feature fusion for rolling bearing fault diagnosis using random forest and autoencoder

To improve the accuracy and robustness of rolling bearing fault diagnosis under complex conditions, a novel method based on multi-view feature fusion is proposed. Firstly, multi-view features from perspectives of the time domain, frequency domain and time-frequency domain are extracted through the Fourier transform, Hilbert transform and empirical mode decomposition (EMD).Then, the random forest model (RF) is applied to select features which are highly correlated with the bearing operating state. Subsequently, the selected features are fused via the autoencoder (AE) to further reduce the redundancy. Finally, the effectiveness of the fused features is evaluated by the support vector machine (SVM). The experimental results indicate that the proposed method based on the multi-view feature fusion can effectively reflect the difference in the state of the rolling bearing, and improve the accuracy of fault diagnosis.