DPT‐tracker:Dual pooling transformer for efficient visual tracking

Transformer tracking always takes paired template and search images as encoder input and conduct feature extraction and target‐search feature correlation by self and/or cross attention operations,thus the model complexity will grow quadratically with the number of input images.To alleviate the burden of this tracking paradigm and facilitate practical deployment of Transformer‐based trackers,we propose a dual pooling transformer tracking framework,dubbed as DPT,which consists of three components:a simple yet efficient spatiotemporal attention model(SAM),a mutual correlation pooling Trans-former(MCPT)and a multiscale aggregation pooling Transformer(MAPT).SAM is designed to gracefully aggregates temporal dynamics and spatial appearance information of multi‐frame templates along space‐time dimensions.MCPT aims to capture multi‐scale pooled and correlated contextual features,which is followed by MAPT that aggregates multi‐scale features into a unified feature representation for tracking prediction.DPT tracker achieves AUC score of 69.5 on LaSOT and precision score of 82.8 on Track-ingNet while maintaining a shorter sequence length of attention tokens,fewer parameters and FLOPs compared to existing state‐of‐the‐art(SOTA)Transformer tracking methods.Extensive experiments demonstrate that DPT tracker yields a strong real‐time tracking baseline with a good trade‐off between tracking performance and inference efficiency.

SiamDLA: Dynamic Label Assignment for Siamese Visual Tracking

Label assignment refers to determining positive/negative labels foreach sample to supervise the training process. Existing Siamese-based trackersprimarily use fixed label assignment strategies according to human priorknowledge;thus, they can be sensitive to predefined hyperparameters and failto fit the spatial and scale variations of samples. In this study, we first developa novel dynamic label assignment (DLA) module to handle the diverse datadistributions and adaptively distinguish the foreground from the backgroundbased on the statistical characteristics of the target in visual object tracking.The core of DLA module is a two-step selection mechanism. The first stepselects candidate samples according to the Euclidean distance between trainingsamples and ground truth, and the second step selects positive/negativesamples based on the mean and standard deviation of candidate samples.The proposed approach is general-purpose and can be easily integrated intoanchor-based and anchor-free trackers for optimal sample-label matching.According to extensive experimental findings, Siamese-based trackers withDLA modules can refine target locations and outperformbaseline trackers onOTB100, VOT2019, UAV123 and LaSOT. Particularly, DLA-SiamRPN++improves SiamRPN++ by 1% AUC and DLA-SiamCAR improves Siam-CAR by 2.5% AUC on OTB100. Furthermore, hyper-parameters analysisexperiments show that DLA module hardly increases spatio-temporal complexity,the proposed approach maintains the same speed as the originaltracker without additional overhead.

In situ atomic-scale observation of size-dependent (de) potassiation and reversible phase transformation in tetragonal FeSe anodes

Potassium-ion batteries(PIBs)are considered promising alternatives to lithium-ion batteries owing to cost-effective potassium resources and a suitable redox potential of-2.93 V(vs.-3.04 V for Li+/Li).However,the exploration of appro-priate electrode materials with the correct size for reversibly accommodating large K+ions presents a significant challenge.In addition,the reaction mecha-nisms and origins of enhanced performance remain elusive.Here,tetragonal FeSe nanoflakes of different sizes are designed to serve as an anode for PIBs,and their live and atomic-scale potassiation/depotassiation mechanisms are revealed for the first time through in situ high-resolution transmission electron micros-copy.We found that FeSe undergoes two distinct structural evolutions,sequen-tially characterized by intercalation and conversion reactions,and the initial intercalation behavior is size-dependent.Apparent expansion induced by the intercalation of K+ions is observed in small-sized FeSe nanoflakes,whereas unexpected cracks are formed along the direction of ionic diffusion in large-sized nanoflakes.The significant stress generation and crack extension originating from the combined effect of mechanical and electrochemical interactions are elucidated by geometric phase analysis and finite-element analysis.Despite the different intercalation behaviors,the formed products of Fe and K_(2)Se after full potassiation can be converted back into the original FeSe phase upon depotassiation.In particular,small-sized nanoflakes exhibit better cycling perfor-mance with well-maintained structural integrity.This article presents the first successful demonstration of atomic-scale visualization that can reveal size-dependent potassiation dynamics.Moreover,it provides valuable guidelines for optimizing the dimensions of electrode materials for advanced PIBs.

A Discussion of Artificial Intelligence in Visual Art Education

Since ChatGPT emerged on November 30, 2022, Artificial Intelligence (AI) has been increasingly discussed as a radical force that will change our world. People have become used to AI in which such ubiquitous technologies as Siri, Google, and Netflix deploy AI algorithms to answer questions, impart information, and provide recommendations. However, many individuals including originators and backers of AI have recently expressed grave concerns. In this paper, the authors will assess what is occurring with AI in Visual Arts Education, outline positives and negatives, and provide recommendations addressed specifically for teachers working in the field regarding emerging AI usage from kindergarten to grade twelve levels as well as in higher education.

3D Object Visual Tracking for the 220 kV/330 kV High-Voltage Live-Line Insulator Cleaning Robot

The 3D object visual tracking problem is studied for the robot vision system of the 220kV/330kV high-voltage live-line insulator cleaning robot. The SUSAN Edge based Scale Invariant Feature (SESIF) algorithm based 3D objects visual tracking is achieved in three stages: the first frame stage,tracking stage,and recovering stage. An SESIF based objects recognition algorithm is proposed to find initial location at both the first frame stage and recovering stage. An SESIF and Lie group based visual tracking algorithm is used to track 3D object. Experiments verify the algorithm's robustness. This algorithm will be used in the second generation of the 220kV/330kV high-voltage live-line insulator cleaning robot.

基于Visual Basic的电子干涉、衍射模拟

根据杨氏双缝干涉、光栅的强度分布函数以及德布罗意物质波的理论,采用蒙特卡罗中的舍选法对大量电子进行舍选,并在Visual Basic中的图片框中输出显示,从而达到在Visual Basic的环境下对电子双缝干涉、光栅衍射的模拟.通过此模拟,可以更加形象地理解电子的波粒二象性以及电子波是概率波,为教学提供方便.

3-D visual tracking based on CMAC neural network and Kalman filter

In this paper, the Kalman filter is used to predict image feature positionaround which an image-processing window is then established to diminish feature-searching area andto heighten the image-processing speed. According to the fundamentals of image-based visual servoing(IBVS), the cerebellar model articulation controller (CMAC) neural network is inserted into thevisual servo control loop to implement the nonlinear mapping from the error signal in the imagespace to the control signal in the input space instead of the iterative adjustment and complicatedinverse solution of the image Jacobian. Simulation results show that the feature point can bepredicted efficiently using the Kalman filter and on-line supervised learning can be realized usingCMAC neural network; end-effector can track the target object very well.

Robust <i>H</i>_∞Observer-Based Tracking Control for the Photovoltaic Pumping System

In this paper, we propose a H∞ robust observer-based control DC motor based on a photovoltaic pumping system. Maximum power point tracking is achieved via an algorithm using Perturb and Observe method, with array voltage and current being used to generate the reference voltage which should be the PV panel’s operating voltage to get maximum available power. A Takagi-Sugeno (T-S) observer has been proposed and designed with non-measurable premise variables and the conditions of stability are given in terms of Linear Matrix Inequality (LMI). The simulation results show the effectiveness and robustness of the proposed method.

An Adaptive Padding Correlation Filter With Group Feature Fusion for Robust Visual Tracking

In recent visual tracking research,correlation filter(CF)based trackers become popular because of their high speed and considerable accuracy.Previous methods mainly work on the extension of features and the solution of the boundary effect to learn a better correlation filter.However,the related studies are insufficient.By exploring the potential of trackers in these two aspects,a novel adaptive padding correlation filter(APCF)with feature group fusion is proposed for robust visual tracking in this paper based on the popular context-aware tracking framework.In the tracker,three feature groups are fused by use of the weighted sum of the normalized response maps,to alleviate the risk of drift caused by the extreme change of single feature.Moreover,to improve the adaptive ability of padding for the filter training of different object shapes,the best padding is selected from the preset pool according to tracking precision over the whole video,where tracking precision is predicted according to the prediction model trained by use of the sequence features of the first several frames.The sequence features include three traditional features and eight newly constructed features.Extensive experiments demonstrate that the proposed tracker is superior to most state-of-the-art correlation filter based trackers and has a stable improvement compared to the basic trackers.

SMSTracker:A Self-Calibration Multi-Head Self-Attention Transformer for Visual Object Tracking

Visual object tracking plays a crucial role in computer vision.In recent years,researchers have proposed various methods to achieve high-performance object tracking.Among these,methods based on Transformers have become a research hotspot due to their ability to globally model and contextualize information.However,current Transformer-based object tracking methods still face challenges such as low tracking accuracy and the presence of redundant feature information.In this paper,we introduce self-calibration multi-head self-attention Transformer(SMSTracker)as a solution to these challenges.It employs a hybrid tensor decomposition self-organizing multihead self-attention transformermechanism,which not only compresses and accelerates Transformer operations but also significantly reduces redundant data,thereby enhancing the accuracy and efficiency of tracking.Additionally,we introduce a self-calibration attention fusion block to resolve common issues of attention ambiguities and inconsistencies found in traditional trackingmethods,ensuring the stability and reliability of tracking performance across various scenarios.By integrating a hybrid tensor decomposition approach with a self-organizingmulti-head self-attentive transformer mechanism,SMSTracker enhances the efficiency and accuracy of the tracking process.Experimental results show that SMSTracker achieves competitive performance in visual object tracking,promising more robust and efficient tracking systems,demonstrating its potential to providemore robust and efficient tracking solutions in real-world applications.

OPTIMIZED MEANSHIFT TARGET REFERENCE MODEL BASED ON IMPROVED PIXEL WEIGHTING IN VISUAL TRACKING

The generic Meanshift is susceptible to interference of background pixels with the target pixels in the kernel of the reference model, which compromises the tracking performance. In this paper, we enhance the target color feature by attenuating the background color within the kernel through enlarging the pixel weightings which map to the pixels on the target. This way, the background pixel interference is largely suppressed in the color histogram in the course of constructing the target reference model. In addition, the proposed method also reduces the number of Meanshift iterations, which speeds up the algorithmic convergence. The two tests validate the proposed approach with improved tracking robustness on real-world video sequences.

Intelligent Biometric Information Management

We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation, and functionality, and robustness refers to the ability to handle incomplete and/or corrupt adversarial information, on one side, and image and or device variability, on the other side. The proposed methodology is model-free and non-parametric. It draws support from discriminative methods using likelihood ratios to link at the conceptual level biometrics and forensics. It further links, at the modeling and implementation level, the Bayesian framework, statistical learning theory (SLT) using transduction and semi-supervised lea- rning, and Information Theory (IY) using mutual information. The key concepts supporting the proposed methodology are a) local estimation to facilitate learning and prediction using both labeled and unlabeled data;b) similarity metrics using regularity of patterns, randomness deficiency, and Kolmogorov complexity (similar to MDL) using strangeness/typicality and ranking p-values;and c) the Cover – Hart theorem on the asymptotical performance of k-nearest neighbors approaching the optimal Bayes error. Several topics on biometric inference and prediction related to 1) multi-level and multi-layer data fusion including quality and multi-modal biometrics;2) score normalization and revision theory;3) face selection and tracking;and 4) identity management, are described here using an integrated approach that includes transduction and boosting for ranking and sequential fusion/aggregation, respectively, on one side, and active learning and change/ outlier/intrusion detection realized using information gain and martingale, respectively, on the other side. The methodology proposed can be mapped to additional types of information beyond biometrics.

Robust visual tracking algorithm based on Monte Carlo approach with integrated attributes

To improve the reliability and accuracy of visual tracker,a robust visual tracking algorithm based on multi-cues fusion under Bayesian framework is proposed.The weighed color and texture cues of the object are applied to describe the moving object.An adjustable observation model is incorporated into particle filtering,which utilizes the properties of particle filter for coping with non-linear,non-Gaussian assumption and the ability to predict the position of the moving object in a cluttered environment and two complementary attributes are employed to estimate the matching similarity dynamically in term of the likelihood ratio factors;furthermore tunes the weight values according to the confidence map of the color and texture feature on-line adaptively to reconfigure the optimal observation likelihood model,which ensured attaining the maximum likelihood ratio in the tracking scenario even if in the situations where the object is occluded or illumination,pose and scale are time-variant.The experimental result shows that the algorithm can track a moving object accurately while the reliability of tracking in a challenging case is validated in the experimentation.

Robust visual tracking for manipulators withunknown intrinsic and extrinsic parameters

This paper addresses the robust visual tracking of multi-feature points for a 3D manipulator with unknown intrinsic and extrinsic parameters of the vision system. This class of control systems are highly nonlinear control systems characterized as time-varying and strong coupling in states and unknown parameters. It is first pointed out that not only is the Jacobian image matrix nonsingular, but also its minimum singular value has a positive limit. This provides the foundation of kinematics and dynamics control of manipulators with visual feedback. Second, the Euler angle expressed rotation transformation is employed to estimate a subspace of the parameter space of the vision system. Based on the two results above, and arbitrarily chosen parameters in this subspace, the tracking controllers are proposed so that the image errors can be made as small as desired so long as the control gain is allowed to be large. The controller does not use visual velocity to achieve high and robust performance with low sampling rate of the vision system. The obtained results are proved by Lyapunov direct method. Experiments are included to demonstrate the effectiveness of the proposed controller.

Morphological disruption and visual tuning alterations in the primary visual cortex in glaucoma(DBA/2J)mice

Glaucoma is a leading cause of irreve rsible blindness wo rldwide,and previous studies have shown that,in addition to affecting the eyes,it also causes abnormalities in the brain.However,it is not yet clear how the primary visual cortex(V1)is altered in glaucoma.This study used DBA/2J mice as a model for spontaneous secondary glaucoma.The aim of the study was to compare the electrophysiological and histomorphological chara cteristics of neurons in the V1between 9-month-old DBA/2J mice and age-matched C57BL/6J mice.We conducted single-unit recordings in the V1 of light-anesthetized mice to measure the visually induced responses,including single-unit spiking and gamma band oscillations.The morphology of layerⅡ/Ⅲneurons was determined by neuronal nuclear antigen staining and Nissl staining of brain tissue sections.Eighty-seven neurons from eight DBA/2J mice and eighty-one neurons from eight C57BL/6J mice were examined.Compared with the C57BL/6J group,V1 neurons in the DBA/2J group exhibited weaker visual tuning and impaired spatial summation.Moreove r,fewer neuro ns were observed in the V1 of DBA/2J mice compared with C57BL/6J mice.These findings suggest that DBA/2J mice have fewer neurons in the VI compared with C57BL/6J mice,and that these neurons have impaired visual tuning.Our findings provide a better understanding of the pathological changes that occur in V1 neuron function and morphology in the DBA/2J mouse model.This study might offer some innovative perspectives regarding the treatment of glaucoma.

Perioperative visual loss after spine surgery

Perioperative visual loss(POVL) is an uncommon, but devastating complication that remains primarily associated with spine and cardiac surgery. The incidence and mechanisms of visual loss after surgery remain difficult to determine. According to the American Society of Anesthesiologists Postoperative Visual Loss Registry, the most common causes of POVL in spine procedures are the two different forms of ischemic optic neuropathy: anterior ischemic optic neuropathy and posterior ischemic optic neuropathy, accounting for 89% of the cases. Retinal ischemia, cortical blindness, and posterior reversible encephalopathy are also observed, but in a small minority of cases. A recent multicenter case control study has identified risk factors associated with ischemic optic neuropathy for patients undergoing prone spinal fusion surgery. These include obesity, male sex, Wilson frame use, longer anesthetic duration, greater estimated blood loss, and decreased percent colloid administration. These risk factors are thought to contribute to the elevation of venous pressure and interstitial edema, resulting in damage to the optic nerve by compression of the vessels that feed the optic nerve, venous infarction or direct mechanical compression. This review will expand on these findings as well as the recently updated American Society of Anesthesiologists practice advisory on POVL. There are no effectivetreatment options for POVL and the diagnosis is often irreversible, so efforts must focus on prevention and risk factor modification. The role of crystalloids versus colloids and the use of α-2 agonists to decrease intraocular pressure during prone spine surgery will also be discussed as a potential preventative strategy.

The Effect of the Menstrual Cycle on Cognitive Performance: Spatial Reasoning, Visual & Numerical Memory

The menstrual cycle has been a topic of interest in relation to behavior and cognition for many years, with historical beliefs associating it with cognitive impairment. However, recent research has challenged these beliefs and suggested potential positive effects of the menstrual cycle on cognitive performance. Despite these emerging findings, there is still a lack of consensus regarding the impact of the menstrual cycle on cognition, particularly in domains such as spatial reasoning, visual memory, and numerical memory. Hence, this study aimed to explore the relationship between the menstrual cycle and cognitive performance in these specific domains. Previous studies have reported mixed findings, with some suggesting no significant association and others indicating potential differences across the menstrual cycle. To contribute to this body of knowledge, we explored the research question of whether the menstrual cycles have a significant effect on cognition, particularly in the domains of spatial reasoning, visual and numerical memory in a regionally diverse sample of menstruating females. A total of 30 menstruating females from mixed geographical backgrounds participated in the study, and a repeated measures design was used to assess their cognitive performance in two phases of the menstrual cycle: follicular and luteal. The results of the study revealed that while spatial reasoning was not significantly related to the menstrual cycle (p = 0.256), both visual and numerical memory had significant positive associations (p < 0.001) with the luteal phase. However, since the effect sizes were very small, the importance of this relationship might be commonly overestimated. Future studies could thus entail designs with larger sample sizes, including neuro-biological measures of menstrual stages, and consequently inform competent interventions and support systems.

Sensor planning method for visual tracking in 3D camera networks

Most sensors or cameras discussed in the sensor network community are usually 3D homogeneous, even though their2 D coverage areas in the ground plane are heterogeneous. Meanwhile, observed objects of camera networks are usually simplified as 2D points in previous literature. However in actual application scenes, not only cameras are always heterogeneous with different height and action radiuses, but also the observed objects are with 3D features（i.e., height）. This paper presents a sensor planning formulation addressing the efficiency enhancement of visual tracking in 3D heterogeneous camera networks that track and detect people traversing a region. The problem of sensor planning consists of three issues：（i） how to model the 3D heterogeneous cameras;（ii） how to rank the visibility, which ensures that the object of interest is visible in a camera＇s field of view;（iii） how to reconfigure the 3D viewing orientations of the cameras. This paper studies the geometric properties of 3D heterogeneous camera networks and addresses an evaluation formulation to rank the visibility of observed objects. Then a sensor planning method is proposed to improve the efficiency of visual tracking. Finally, the numerical results show that the proposed method can improve the tracking performance of the system compared to the conventional strategies.

Visual Quality Analysis of Urban Park Landscapes Based on Eye Tracking: A Case Study of Nanjing Xiaohong Stone Carving Park

The landscape quality of urban parks is an important aspect of tourists'landscape experience.A full understanding of their relationship is beneficial to the park design,construction and management.In this paper,Xiaohong Stone Carving Park in Nanjing was selected as a case study.Firstly,the subjective landscape visual quality evaluation of tourists was obtained through the scenic beauty evaluation,and the objective landscape visual preference of users was obtained through eye movement experiment and analysis.Secondly,the values of various park landscape elements were measured respectively.And then the correlation between the subjective and objective evaluation and the values of park landscape elements was analyzed.The results showed that:(1)The visual preference of plant cluster landscape with beautiful shape,distinct layers and significant color contrast is high;(2)The preference for waterfront scenes with significant terrain differences,rich background vegetation levels,and clear and vivid water side reflections is high;(3)The designed artificial structures often become the focus of attention in the whole scene,which can improve the beauty of the scene;(4)The overall landscape composition is very important for the beauty of the scene.Finally,through the comprehensive analysis with the design expectation,the optimization strategy was proposed.

Visual Prolog的搜索控制机制分析

回溯机制是逻辑程序设计的重要设施。回溯本身是一种获得目标所有可能解的良好方法。然而回溯也有副作用,一是它可能导致Visual Prolog给出多余的答案,而Visua1 Prolog自己不能区分实质上相同的两个解,因此会降低效率;二是尽管一个特殊的目标已被满足,但是回溯机制可能还会强迫Visual Prolog 继续寻找另外的解,因此会增加系统开销。在这些情况下,必须仔细控制目标搜索求解的回溯过程。本文在揭示Visual Prolog回溯机制所存在问题的基础上,通过实例,对Visual Prolog的静态截断机制、失败谓词fail与否定谓词not等控制谓词,以及动态截断机制等所构成的完整的目标搜索求解控制机制进行了详细分析,从而揭示出回溯机制和搜索求解控制机制的本质特性及应用机理。