Simulator sickness has been one of the major obstacles toward making virtual reality (VR) widely accepted and used. For example, navigation produces vection, which is illusion self-motion as perceives bodily motion despite no movement actually occurs. This, in turn, causes a sensory conflict between visual actual (or vestibular) sickness. In this study, we explore method to reduce simulator by visually mixing optical flow patterns that are reverse direction motion. As mainly detected...

This study investigates how different virtual agent (VA) behaviors influence subjects' perceptions and group decision-making. Participants carried out experimental discussions with a VA exhibiting varying levels of engagement affective behavior. Engagement refers to the VA's focus on task, whereas behavior reflects emotional state. The findings revealed that engagements effectively captured participants' attention even in setting enhanced synergy, thereby facilitating more in-depth...

This poster introduces the Masked FER-2013 dataset that can be used to analyze facial emotions in people wearing masks and train a recognizer. The containing face images annotated with seven was modified by synthesizing mask into lower portion of face. Based on quantitative evaluation, our improve accuracy CNN, VGG, ResNet models' emotion recognition masked maximum 46%. Additionally, as use case, we present an application recognizes person, generates emotional animation, augments it real...

AudienceMR is designed as a multi-user mixed reality space that seamlessly extends the local user to become large, shared classroom where some of audience members are seen seated in real space, and more through an extended portal. can provide sense presence large-scale crowd/audience with associated spatial context. In contrast virtual (VR), however, (MR), lecturer deliver content or conduct performance from real, actual, comfortable, familiar while interacting directly nearby objects, such...

摘要: 该文针对K平面聚类算法KPC (K-Plane Clustering)对噪声点敏感的缺陷，通过引入隶属度约束函数，推导出鲁棒的改进分割K平面聚类算法IFP-KPC(Improved Fuzzy Partitions for K-Plane Clustering)，并利用Voronoi距离对IFP-KPC算法的鲁棒性进行了合理解释。实验结果表明IFP-KPC算法较之于KPC算法具有更好的聚类效果。 关键词: K平面聚类; 改进模糊分割; Voronoi距离; 鲁棒性

Armored equipment plays a crucial role in the ground battlefield. The fast and accurate detection of enemy armored targets is significant to take initiative Comparing general object vehicle detection, target battlefield environment more challenging due long distance observation complicated environment. In this paper, an robust automatic method proposed detect Firstly, inspired by Feature Pyramid Network (FPN), we propose top-down aggregation (TDA) network which enhances shallow feature maps...

In this paper, a fragile audio watermarking scheme for content authentication is proposed. The signal segmented and the sum of absolute value samples computed each part, then frame adjusted quantized quantization result converted to binary as watermark. Then, proper positive integer chosen compute mean continuous sample amplitude by sliding one every time. sequence constructs line which cross timing axis in time domain. Watermark information embedded at these intersection points with dither...

A robust speech content authentication algorithm against desynchronization attacks is proposed, in order to authenticate the of digital signal. Firstly, signal framed, and each frame divided into two parts. The number frame, as watermark bit, embedded first second part by quantizing Bessel-Fourier moments correspond signals. authenticated comparing with bits extracted from part. Experiment results shown that scheme proposed effective. And some existing schemes, performance improved. Index...

We discuss how data fusion and mining are mutually supportive in multisensor systems and, with an illustrative example, we show the rough set theory is a powerful tool to inducing rule-based object models for system used environment pollution monitoring. The entropy employed analyze attribute sets. Also, probability model introduced mine statistical information hidden table. They all promising future processing coming from system.

In order to identify Tor anonymous network traffic which was generated by the most widely used in world, analyzing features that can be recognize based on meek pluggable transport, and proposing a method machine learning classify traffic. identification aimed at Tor-Meek using Meek confusion technology network. To determine flow characteristics from original feature set, RandomForest selection is evaluate importance of these features, select available subset. The classifier constructed C4.5,...

As a software representation of assets and processes, digital twins is valuable for the design future marine monitoring control systems. Such system usually requires integration technologies from cloud edge, WebAccess/SCADA SaaS Composer Advantech Technology were adopted in our proposed framework ships running coastal waters based application was given paper. The compatible with various network access technologies, including near shore oriented cellular communication off satellite...

In this paper, a new reduct searching algorithm is proposed to benefit the applications of rough sets theory. The information entropy introduced algorithm, so that indeterministic causalities among attributes can be found and sensitivity noise, occurring unavoidably when approximation quality function /spl gamma/ applied, removed. theoretical analysis an illustrated example show rule set induced by uses less than algorithms based on function. At same time, gives larger coverage, especially...

Noise is often brought to host audio by common signal processing operation, and it usually changes the high-frequency component of an signal. So embedding watermark adjusting low-frequency coefficient can improve robustness a scheme. Moving Average sequence feature This work proposed method which into maximal in discrete cosine transform domain moving average sequence. Subjective objective tests reveal that watermarking scheme maintains highly quality, simultaneously, algorithm robust...

The synchronization scheme based on moving average is robust and suitable for the same rule to be adopted in embedding watermark code, but imperceptibility search efficiency seldom reported. study aims improve original audio watermarking. Firstly, survival of algorithm from desynchronization attacks improved. Secondly, improved inaudibility. Objective difference grade (ODG) marked significantly changed. Thirdly, analyzed derived result close experimental result. Fourthly, selection...

In this work, we propose to use an approximate 2D map of the environment generated from latest modeling technology and enhance object manipulation performance for touch based mobile augmented reality. We validated advantage proposed interface through a pilot experiment confirmed that helps reduce task completion time almost 2 times improve usability as well.

In light of the COVID-19 pandemic, wearing a mask is crucial to avoid contracting infectious diseases. However, known impair communication functions. This study aims address difficulties caused by and provide strategy for aiding in understanding speaker's speech through facial animation. Facial animation generated real-time, upper information processed detect emotions, generating lower expression. addition, system detects mask's shape enables accurate registration proper position. technology...

Serendipity is the phenomenon of people making unexpected and beneficial discoveries. While research on mechanism effectiveness serendipity in information acquisition has been actively conducted, little attempt made to quantify when it occurs. In this paper, we present a quantitative model that measures experienced by users hypertext environment. order propose an evaluation how probable would experience serendipitous moments process active search, define discovery as can happen during...

We propose an intrusion detection system for wireless sensor networks. According to the features of clustered network, in every node network deploy IDS agent, which include two different agent: local agent and global agent. This carry out tasks, at same time using several blocks order realize data collection, detection, response agents management, would make our has advantages simple operation good expansibility, lower cost improved secure capacity.