你也来当一回画家

每次看到蒙娜丽莎、向日葵、日出印象这些名画时，除了对着画框向艺术大师表示敬意外，是否也想过自己挥舞画笔，指点油彩、激扬画布？可惜知道自已没有那份才能。在电脑上倒是可以用PS做出类似特效来，但是往往复杂难学。难道这辈子就与绘画无缘？其实只要利器在手，你会发现原来这个梦想很容易实现。     

基于模糊理论的人体动作表示及动作模式识别方法

根据模糊数学理论和模糊神经网络技术,本文提出了一种适用于体育动作分析的人体动作元层次结构模型、动作特征的模糊表示方法和体育动作模糊识别方法,并给出了实现体育动作模糊识别的过程和实验结果.     

基于实时视频感知的虚拟体育交互系统

针对疫情常态化背景下,传统体育项目受场地、器材等限制,市场上相关产品价格昂贵、可扩展性不足等问题,提出了一种基于实时视频感知的虚拟体育交互系统.该系统设计视频数据采集模块和人体关节点提取模块,结合OpenPose获取人体的关节点坐标,实时捕捉人体手势以及肢体动作.动作语义理解模块包括运动动作理解和绘图动作理解.前者根据运动中肢体关节点的相对位置关系,识别运动动作语义.后者将手腕部关节点绘图动作轨迹生成为草图图像,使用AlexNet进行识别分类,解析为对应的绘制动作语义.该模型在边缘端设备的分类准确率为98.83%.采用基于Unity设计的草图游戏应用作为可视化交互界面,实现在虚拟场景中的运动交互.该系统使用实时视频感知交互方式实现居家运动健身,无需其他的外部设备,具有更强的参与度和趣味性.     

基于时域分割的人体行为连续性动作预测仿真

传统人体行为动作预测方法对人体骨架上的所有关键特征映射点包含范围模糊,导致实时处理速度慢,无法有效对连续性动作进行预测,且错位率较高.为此提出一种基于时域分割的人类行为连续性动作预测.首先通过三维空间捕捉技术构建人体骨架模型,通过时域分割确定关节点的位置信息,将任意关节点都视为映射特征点,提取人体的运动数据特征,利用LLE把提取的特征数据映射进二维空间里,结合拉格朗日的乘法最优化重建权值矩阵,建立动作向量库,确定人体做出随意性动作与向量库中动作相似或者相同时,就能够对人体的动作进行提前预判,即可以完成动作预测.仿真结果证明,所提方法时域分割效果好,实时预测速度快,能够有效的对人体连续性动作精准预测.     

基于时序深度置信网络的在线人体动作识别

在线人体动作识别是人体动作识别的最终目标,但由于如何分割动作序列是一个待解决的难点问题,因此目前大多数人体动作识别方法仅关注在分割好的动作序列中进行动作识别,未关注在线人体动作识别问题.本文针对这一问题,提出了一种可以完成在线人体动作识别的时序深度置信网络（Temporal deep belief network, TDBN）模型.该模型充分利用动作序列前后帧提供的上下文信息,解决了目前深度置信网络模型仅能识别静态图像的问题,不仅大大提高了动作识别的准确率,而且由于该模型不需要人为对动作序列进行分割,可以从动作进行中的任意时刻开始识别,实现了真正意义上的在线动作识别,为实际应用打下了较好的理论基础.     

光学动作捕捉中缺失标记重建方法研究

光学动作捕捉技术是一种常用的动作捕捉方法,目前已经在各个行业内广泛应用.尤其是在体育竞技领域,已经成为了不可或缺的训练辅助手段.在光学动作捕捉中,最常见的问题是缺失标记,可能由外部遮挡、身体自遮挡或信号丢失等原因造成.对于缺失标记问题,在以往的研究中要么需要舍弃缺失的标记,要么需要大量后处理工作来恢复缺失标记.针对这种情况,本文提出一种用于光学动作捕捉中缺失标记的重建方法,该方法使用卡尔曼滤波框架,结合运动数据来预估缺失标记点的位置,实时重建人体运动模型.实验结果证明该方法能够快速有效的恢复缺失标记,重建人体运动.     

基于雷达传感器的连续人体动作识别方法

针对基于雷达传感器的离散人体动作识别方法难以得到实际应用的问题,文章提出了一种基于雷达传感器的连续人体动作识别方法。首先对连续动作的雷达回波信号进行预处理得到距离时间域图像。然后通过时频分析得到微多普勒时频谱图像。最后分别采用支持向量机与长短期记忆网络作为分类器进行动作识别。实验结果表明,采用长短期记忆网络作为分类器对人体连续动作进行分类的平均识别准确率达到80.7%,对比SVM作为分类器得到的平均识别准确率高4%。     

足球机器人交互规则控制系统设计

在足球机器人人机交互运动辨识中,由于人类在足球运动中复杂的姿势、行为动作、生理心理状态、语言、情感和触觉等自然能力,造成了机器人在对足球运动动作辨识过程中存在大量的伪动作指令.传统的人机交互动作辨识方法去除伪动作所应用的计算过程过于复杂,致使识别速度缓慢,效率低.提出一种新的足球机器人人机交互运动辨识方法,选择中型组足球机器人UP-VoyagerⅢ作为研究对象,应用Kinect体感技术实现人体骨骼的检测和跟踪,设计人体前进、后退、左行、右行和停止等上肢动作规则.以中型组足球机器人持球器方向为正前方,针对对应规则设计圆周等距排列的3轮组成万向运动底层速度分解策略,以实现快速的人机交互功能.实验表明该中型组足球机器人可以根据系统架构实现人体动作辨识,进而执行相应动作.研究方法为中型组机器人辨识人动作信息提供了实用和可靠的方法.     

提示笔刷的数字油画画布自动生成

传统的数字油画画布由区域边界线条和指定颜料数字构成,绘画者仅需在每个区域内部涂上与数字对应的颜料即可完成自己的手绘作品,是一种流传广泛的新艺术形式。然而,这些画布不仅需要专业人员花费大量时间设计,且无法提示笔刷的方向和大小,影响绘画的效果。为此,提出一种提示笔刷的自动生成方法,将任意图像转换为数字油画画布的同时,利用方向场流线合理提示出笔刷的宽度、长度和方向。实验结果表明,该方法生成的数字油画画布线条封闭光滑,数字清晰可辨,笔刷提示流线对绘画效果有明显的改善。     

Wi-Do:WiFi信号下的高鲁棒人员动作感知模型

人机交互是物联网迈向智能化的重要途径,而人体动作识别已成为智能环境实现的关键环节.由于WiFi具有良好的用户体验和极高的普适性以及低廉的部署成本,基于WiFi的人体运动识别技术从众多交互技术中脱颖而出,已在智能安防、运动保健、老年活跃检测等领域展现了巨大的应用价值.现有的WiFi动作识别工作中,动作识别受人体的运动方向影响严重,为了确保识别精度往往需要固定动作方向,这种方向依赖性对基于WiFi的动作识别技术造成了极大的阻碍.为了克服这一限制,提出一种方向无关的动作识别模型.该模型利用天线分集消除随机的相位偏移,将人体运动在频域上造成的多普勒频移与快速傅里叶变换值作为识别特征,并引入注意力机制的双向GRU(gate recurrent unit)来对运动进行分类识别.该模型将空间特征集成到时间模型中,提升了无线信号对人体动作识别的鲁棒性与准确性.在典型室内环境下的实验结果显示了优越的性能与93%的准确率,验证了该模型优于之前的识别模型.     

A brush device with visual and haptic feedback for virtual painting of 3D virtual objects

We have previously developed a mixed reality (MR) painting system with which a user could take a physical object in the real world and apply virtual paint to it. However, this system could not provide the sensation of painting on virtual objects in MR space. Therefore, we subsequently proposed and developed mechanisms that simulated the effect of touch and movement when a brush device was used to paint on a virtual canvas. In this paper, we use visual and haptic feedback to provide the sensation of painting on virtual three-dimensional objects using a new brush device called the MAI Painting Brush++. We evaluate and confirm its effectiveness through several user studies.     

A virtual 3D interactive painting method for Chinese calligraphy and painting based on real-time force feedback technology

A novel 3D interactive painting method for Chinese calligraphy and painting based on force feedback technology is proposed. The relationship between the force exerted on the brush and the resulting brush deformation is analyzed and a spring-mass model is used to build a model of the 3D Chinese brush. The 2D brush footprint between the brush and the plane of the paper or object is calculated according to the deformation of the 3D brush when force is exerted on the 3D brush. Then the 3D brush footprint is obtained by projecting the 2D brush footprint onto the surface of the 3D object in real time, and a complete 3D brushstroke is obtained by superimposing 3D brush footprints along the painting direction. The proposed method has been successfully applied in a virtual 3D interactive drawing system based on force feedback technology. In this system, users can paint 3D brushstrokes in real time with a Phantom Desktop haptic device, which can effectively serve as a virtual reality interface to the simulated painting environment for users.     

Painting robot with multi-fingered hands and stereo vision

In this paper, we describe a painting robot with multi-fingered hands and stereo vision. The goal of this study is for the robot to reproduce the whole procedure involved in human painting. A painting action is divided into three phases: obtaining a 3D model, composing a picture model, and painting by a robot. In this system, various feedback techniques including computer vision and force sensors are used. As experiments, an apple and a human silhouette are painted on a canvas using this system.     

Hardware‐Based Non‐Photorealistic Rendering Using a Painting Robot

We describe a painting machine and associated algorithms. Our modified industrial robot works with visual feedback and applies acrylic paint from a repository to a canvas until the created painting resembles a given input image or scene. The color differences between canvas and input are used to direct the application of new strokes. We present two optimization‐based algorithms that place such strokes in relation to already existing ones. Using these methods we are able to create different painting styles, one that tries to match the input colors with almost transparent strokes and another one that creates dithering patterns of opaque strokes that approximate the input color. The machine produces paintings that mimic those created by human painters and allows us to study the painting process as well as the creation of artworks.     

Interactive digital graffiti canvas system

This paper presents an interactive digital painting system that allows a user to draw graffiti on a virtual 3D canvas with a digital spray can. The system visualizes a stereoscopic representation of the canvas by tracking the user's head. It also emulates real‐time spray painting by tracking the spray can in the user's hand as well as sensing the button pressure of the spray device. After painting a 3D object, the user can interact with the object on the display and see it flying in the 3D environment through a tracked head‐mounted display. As demonstrated in the results of our evaluation, we verified that the system resembles real graffiti in regard to a natural and realistic graffiti experience. Copyright © 2015 John Wiley & Sons, Ltd.     

非真实感油画绘制中明暗色调获取方法研究

作为与真实感图形学相对应的图形学分支,近年来非真实感图形学的发展受到人们的关注并已成为计算机图形学的研究热点之一.然而,非真实感绘制作为一个新的研究领域在很多方面还处于不断发展之中.在非真实感油画的绘制中,场景中明暗色调的合理性和协调性直接影响到最后生成画面的效果.对非真实感油画绘制中明暗色调的获取方法进行研究,提出一种基于小平面片的色调信息提取方法.该方法假设场景中客体表面均匀覆盖了一层大小相同的正方形"小平面",这些小平面起到了"提取器"的作用.通过计算这些小平面被遮挡的明暗信息和近似的镜面反射信息等来确定小平面所在位置的色调信息,在绘制过程中利用这些色调信息在相应位置上绘制"笔触",从而保证了所生成油画画面场景整体色调的合理性.所提出的方法结构简单、计算速度快,同时可方便实现对场景中局部信息的调整.实验结果验证了所提出方法的有效性.     

The art of interaction

Interactive art has become much more common as a result of the many ways in which the computer and the Internet have facilitated it. Issues relating to human–computer interaction (HCI) are as important to interactive art making as issues relating to the colours of paint are to painting. It is not that HCI and art necessarily share goals. It is just that much of the knowledge of HCI and its methods can contribute to interactive art making. This paper reviews recent work that looks at these issues in the art context. In interactive digital art, the artist is concerned with how the artwork behaves, how the audience interacts with it and, ultimately, in participant experience and their degree of engagement. The paper looks at these issues and brings together a collection of research results and art practice experiences that together help to illuminate this significant new and expanding area. In particular, it is suggested that this work points towards a much needed critical language that can be used to describe, compare and discuss interactive digital art.     

Making it new: transformations in the grammar and identity of digital printmaking

Notions of skill and sensitivity in fine art printmaking have been transformed by the new action and knowledge environments brought about by digital technologies. However, the off-the-shelf programs used in these environments are still influenced by an ideology that promotes the simulation of traditional techniques, which in turn prevent a move away from conventional forms of depiction. The ideological stance that drives these programs is problematic because it re-enforces out-dated artistic paradigms. In order to move beyond these paradigms, artists who work with digital technologies will need to be more critical of attempts merely to import techniques analogous to those used in traditional practice (e.g. processing applications that only provide a virtual equivalent to traditional painting, drawing, montage, and photographic methods) and will have to face down the tendency to mimic existing paint and print media. What we need for computer printmaking is the sort of interdependence and autonomy gained by an all-digital approach.     

SIAS, Strokes Interpreted Animated Sequences

The goals of the bulk of commercial computer systems for graphics and animated film/video are different from the kind of systems involving the production of art. The former systems are developed for industrial application. Accurate technical communication and precise representation of measures are the main goals of the drawings and die coloured plates intended for industrial application.
In the arts, however, the main goal of a drawing, a painting, or an animated film/video is beauty. The visual artist searches for an equilibrium between proportions of different shapes and an equilibrium between different colours. A system able to help in this search must therefore have its roots in art, i.e. the so-called modern art. Starting from my experience in painting, in animated film by pencil drawing, and in computer animation, I concluded that a scene could be represented very effectively and accurately by the technique of painting and drawing.
The system presented for automatic painting of scenes within a 3D space that are changing in time, is based on the tradition of drawing and painting. The system must be applicable within the visual arts and video production. An experimental computer animation system SIAS (Strokes Interpreted Animated Sequences) has been developed which produces coloured shapes that are similar to the strokes put on a canvas by a painter.     

虚拟现实在体育仿真中的应用综述

对基于虚拟现实的体育(训练)仿真技术的发展现状进行综述，讨论了体育仿真的关键技术，介绍了几个典型应用系统，最后对虚拟现实技术在体育训练仿真中的应用前景进行了讨论。