Augmenting video surveillance footage with virtual agents for incremental event evaluation

The fields of segmentation, tracking and behavior analysis demand for challenging video resources to test, in a scalable manner, complex scenarios like crowded environments or scenes with high semantics. Nevertheless, existing public databases cannot scale the presence of appearing agents, which would be useful to study long-term occlusions and crowds. Moreover, creating these resources is expensive and often too particularized to specific needs. We propose an augmented reality framework to increase the complexity of image sequences in terms of occlusions and crowds, in a scalable and controllable manner. Existing datasets can be increased with augmented sequences containing virtual agents. Such sequences are automatically annotated, thus facilitating evaluation in terms of segmentation, tracking, and behavior recognition. In order to easily specify the desired contents, we propose a natural language interface to convert input sentences into virtual agent behaviors. Experimental tests and validation in indoor, street, and soccer environments are provided to show the feasibility of the proposed approach in terms of robustness, scalability, and semantics.     

Investigating stacking: How do registered nurses prioritize their activities in real-time?

Recent nursing graduates often find it challenging to prioritize competing tasks in the increasingly complex hospital environment. In order to develop a normative framework for task prioritization, Registered Nurses from a variety of settings were interviewed about how they prioritized their activities in the workplace. From 30 interviews, 422 prioritization decisions were coded when study participants were confronted with two tasks that could not be done simultaneously. The findings suggest a seven-level prioritization hierarchy of nursing activities: 1) addressing imminent clinical concerns, 2) high uncertainty activities, 3) significant, core clinical caregiving and managing pain, 4) relationship management, 5) documenting, helping others, and patient support, 6) system improvement and cleaning/preparing supplies, and 7) personal breaks and social interactions. Explicitly providing a normative framework for task prioritization during nursing education may help to accelerate learning this skill. For more experienced nurses, we believe that flexibility to deviate from this framework will be important for providing high-quality, personalized care that best matches the particular context.

Relevancy to industry

Nursing personnel comprises a significant portion of the human capital and operating budget for hospitals. Knowledge of what tasks are likely to be shed under severe workload conditions aids with optimizing hiring and allocation of nursing personnel.     

A Primer to Human Threading

Human Threading™ is new approach in developing innovative computing technologies. It uses novel physiologic combinations to measure the human brain and body in an effort to create greater efficiency among human and machine. A divergent group of measurement techniques are employed in this process including EEG, NIRS, fMRI, GSR, Saccade testing and more. An overarching goal in this area of study will be the development of simultaneous, multi-modal operating platforms providing efficiency and interchangeability. As an example this paper provides a comparison of efficiency between two email systems. The first system is a standard Microsoft Outlook-based email application using Microsoft XP. The next is a voice-based email system created at the Human Threading™ research laboratory. EEG and time trials were performed on ten participants with a total of 30 trails. Voice-based email was shown to be 220% faster and use statistically less cognitive load on participants prefrontal cortex. This paper shows initial results in support of Human Threading™ and the potential power behind efficient, modal specific operating platforms.     

Predictive models of human supervisory control behavioral patterns using hidden semi-Markov models

Behavioral models of human operators engaged in complex, time-critical high-risk domains, such as those typical in Human Supervisory Control (HSC) settings, are of great value because of the high cost of operator failure. We propose that Hidden Semi-Markov Models (HSMMs) can be employed to model behaviors of operators in HSC settings where there is some intermittent human interaction with a system via a set of external controls. While regular Hidden Markov Models (HMMs) can be used to model operator behavior, HSMMs are particularly suited to time-critical supervisory control domains due to their explicit representation of state duration. Using HSMMs, we demonstrate in an unmanned vehicle supervisory control environment that such models can accurately predict future operator behavior both in terms of states and durations.     

BPEL中基于异步模式的人工任务执行系统研究

BPEL(Business Process Execution Language)是一种编写基于Web服务的自动化业务流程的语言,但不支持用户交互,用户交互可以通过人工任务实现。为此,提出一种在BPEL中支持人工任务执行的系统架构。在引擎之外定义一个人工任务管理器来维护人工任务,流程与人工任务管理器采用异步消息模式进行交互,以便更适应人工任务执行时间不确定的特点。使用13PEI提供的消息相关集关联异步交互过程中调用人工任务管理器的活动与接收人工任务管理器执行结果的活动。     

基于人类视觉系统的数字图像水印方法

基于人类视觉系统(HVS)提出一种可有效平衡数字水印鲁棒性和不可见性的方法。利用基于矩的图像归一化技术将载体图像映射到几何不变空间内,并结合不变质心理论在归一化图像中确定重要区域。根据人眼视觉系统掩蔽特性将一个大小为256 bit的可读水印自适应地嵌入到重要区域的离散傅里叶变换(DFT)域内,同时采用预失真补偿策略弥补图像归一化误差较大的缺陷。仿真实验结果表明,该算法不仅具有良好的透明性,而且对常规信号处理和大部分几何攻击均具有较强的鲁棒性。     

基于Kinect的高维人体动画实时合成研究

基于动作的视频交互游戏一直是游戏市场上非常受消费者青睐的主流游戏之一。研究利用Kinect从用户动作中获取低维控制信号,然后通过双级结构来重建高维度运动控制信号,以实现高质量人体动画的实时合成。其中第一级先通过构造一个邻居图缩小搜索空间,再通过K-D树加速搜索算法得到k个相似数据,最后基于主成分分析法来构建一个线性运动实时合成模型;第二级则是利用平滑参数对线性模型进行优化。实验结果表明,即使在场景受到严重干扰的情况下,该方法仍然可以重建出高质量的人体动画。     

基于重建编码信息的视频ROI提取算法

针对H.264/AVC视频编码标准计算复杂度高、难以实时应用的缺陷,利用重建块的运动矢量和帧间预测模式,并结合人类视觉感知特征,提出一种视频感兴趣区域快速提取算法及其优化编码方案。仿真结果表明,与H.264/AVC算法相比,该算法能够在码率平均增加0.17%、峰值信噪比平均下降0.052 dB的前提下,节省53.11%的编码时间。     

车间数字化制造模式和Web服务移动控制数控系统研究

在中国国家自然科学基金项目《人机协同的车间数字化制造模式——基于移动通信的车间数字助理和掌上数据研究》资助下， 以制造车间为研究对象，开展人机协同的车间数字化制造模式及其关键技术研究，对其成果进行学术总结。 提出车间数字化制造(Digital Shop Manufacturing，DSM)概念。DSM以制造科学技术为基础，应用数字技术对车间制造中所涉及的所有对象和活动给予表示、进行处理和控制，具有信息化、网络化、智能化、自动化等特点。提出数字化车间系统(Digital Shop System，DSS)概念，将DSS分为车间数字化管理软件系统(Shop Digital Management Software System，SDMSS)和车间控制系统(Shop Control System，SCS)等两部分。     

浅谈企业的科技人才管理

企业的科技人才管理,是企业人力资源管理的重要组成部分,也是科技管理的重要内容.文章从企业的组织制度体系、开发培训、激励等方面,分析了我国企业科技人才管理的现状及存在的问题,提出了企业科技人才的管理创新方案.