基于Leap Motion的动态手势识别

动态手势识别作为人机交互的一个重要方向，在各个领域具有广泛的需求。相较于静态手势，动态手势的变化更为复杂，对其特征的充分提取与描述是准确识别动态手势的关键。为了解决对动态手势特征描述不充分的问题，利用高精度的Leap Motion传感器对手部三维坐标信息进行采集，提出了一种包含手指姿势和手掌位移的特征在内的、能够充分描述复杂动态手势的特征序列，并结合长短期记忆网络模型进行动态手势识别。实验结果表明，提出的方法在包含16种动态手势的数据集上的识别准确率为98.50%；与其他特征序列的对比实验表明，提出的特征序列，能更充分准确地描述动态手势特征。     

Leap motion controller three dimensional verification and polynomial correction

We propose a leap motion controller (LMC) dimensional verification based on ISO 10360-2:2009 with a coordinated measuring machine (CMM) as the reference framework. A pointer device comprising a thin aluminum cylinder was used to simulate a human finger. This was mounted on the spindle of the CMM to mark known positions over the LMC workspace. Polynomial tendency line corrections were applied to reduce the error in the LMC and CMM framework alignment. One dimension verification results were less than 0.1 mm in the X and Z axes, whereas the Y axis produced unsuitable results. The mean error was 9.6 mm in three-dimensional (3D) verification. Our findings demonstrate the difference between manufacturer quoted accuracy (0.01 mm) to that practically obtainable when the pointer was placed in a known position. LMC needs to add tracking models and position error compensation in applications requiring high accuracy, such as industrial processes or surgical procedure simulations.     

Feasibility of tracking laparoscopic instruments in a box trainer using a Leap Motion Controller

Motion analysis is employed to assess minimally invasive surgical psychomotor skills in box trainers. Tracking of laparoscopic instruments requires sensor-based systems that can be expensive, limit movements and modify their ergonomic properties. We evaluate the feasibility of using Leap Motion as a cheap, unobtrusive alternative. Four experiments were performed to determine its precision while tracking a laparoscopic instrument inside and outside a box trainer. Static long and short term precision of the Leap Motion was <2.5 mm. Precision between 12 different positions within the box trainer was <0.7 mm for all measured distances between neighbors. Dynamic precision when moving the instrument for 200 mm ranged between 2 and 15 mm. Leap Motion presents acceptable precision values inside a box trainer. Improvements are still required (e.g.: multiple instruments’ tracking). Once solved, a validation study should verify the usefulness of Leap Motion to objectively measure skills of novices and residents during training.     

马克思跨越论的设想及其方法论意义

马克思的跨越论只是一种假设,而不是一个肯定的科学结论;它只是提出问题,而不是最后解决了问题。马克思关干俄国跨越资本主义“卡夫丁峡谷”设想的意义并不在这一设想本身。而是在于这一设想为我们提供了研究落后国家社会发展道路的科学方法论,即生产力与生产关系矛盾运动的民族性和世界性相互作用的辩证法。     

跳跃式经济发展模式及其路径选择

根据新兴古典经济学的分工理论和制度学派的理论工具 ,论述了跳跃式经济发展的两种大推进工业化模式 ,以及由制度结构变迁引发的跳跃式经济发展模式 ,并提出了跳跃式经济发展模式的实现路径     

基于手势的自然用户界面(NUI)在会议中的应用

自然用户界面NUI的清晰、简洁、智能化的方式让用户体验更加直观,更为人性化。基于自然用户界面的应用设备层出不穷,手势识别这种操作技术占据了重要的位置。随着虚拟现实技术快速发展,人们的办公方式也发生了很大的变化。在幻灯片演示方面,传统的方式显得呆板,不能让人们自由使用动作。因此,文中提出一种使用手势来代替鼠标输入的方法,使用Leap Motion控制器。经过试验,计算机的反应速度快,并能准确地辨别出有效手势,能够对幻灯片进行操作。     

The concept of leap in measurement interpretation

This paper discusses a problem in the way people think about measurement. This problem is analyzed using the tool of an Object Oriented Model of Measurement, developed in the paper, which is related to the Representational Theory of Measurement but includes alters or extends the tenets of that theory. It is concluded that all measurement involves a leap of decision concerning the model of analysis of the observed which is to be employed, and in addition information processing often involves a leap from the algorithmically derivable to useful conclusions about the observed as sought through the measurement process.     

基于Leap Motion手语语音转换的设计与实现

文章介绍了基于Leap Motion手语语音转换的原理以及在实际中如何实现应用,其能够将手语手势翻译成所需语言,并且以语音的形式传递消息,实现为聋哑人"发声"的目的。Leap Motion手语语音转换的硬件部分采用Leap Motion体感控制器,能够实时采集人手指的坐标、加速度、方向等动态手势的运动信息。软件采用Unity 3D三维引擎,提供了开发手势识别的SDK,并给出了一些基本手的模型,在Unity 3D中完成由手势到文字,再到语音的转换。     

基于小波分析的白车身阶跃测量数据处理

白车身焊装质量评价过程中，三坐标测量机(CMM)测量数据为非平稳随机过程。如何对白车身焊装质量做出准确评价是对其制造质量进行控制的基础。本文采用小波分析方法对白车身焊装过程中具有阶跃特征的CMM测量数据进行处理，使用处理后的数据对白车身焊装质量进行评价，并对数据处理过程中的问题进行了研究。通过对一个实际案例的应用分析，验证了小波分析方法对白车身阶跃测量数据处理的有效性。     

基于Leap Motion的动态手势识别方法

伴随虚拟现实（Virtual Reality,VR）技术的发展,以及人们对人机交互性能和体验感的要求提高,手势识别作为影响虚拟现实中交互操作的重要技术之一,其精确度急需提升^[1].针对当前手势识别方法在一些动作类似的手势识别中表现欠佳的问题,提出了一种多特征动态手势识别方法.该方法首先使用体感控制器Leap Motion追踪动态手势获取数据,然后在特征提取过程中增加对位移向量角度和拐点判定计数的提取,接着进行动态手势隐马尔科夫模型（Hidden Markov Model,HMM）的训练,最后根据待测手势与模型的匹配率进行识别.从实验结果中得出,该多特征识别方法能够提升相似手势的识别率.