Abstract
Interactive animation based on motion capture is a relatively new animation technique that implements the action of transforming an actor’s actions into a model within the computer in real time. To solve this problem, a visualized interactive computer vision system was developed. A brief introduction to the design of the entire system and some important technologies, focusing on the principles of node-based visual programming, the integration of functional modules, and the plug-in architecture of the system. The experimental results show that the process is simple, fast and real-time, especially in the field of entertainment and education.
Introduction
With the development of the times, the emergence of the word “interaction” must be accompanied by the birth of the visual interaction design industry. The interactive model has emerged in graphic design, landscape design, interior design, and animation design. “Interaction” from the initial interaction between people, to the interaction between people and machines today, from the hard acceptance of the beginning to the active participation and interaction now, the interaction has gradually penetrated into people’s real life. Its humanized design allows us to enjoy the changes in the sensory world and lifestyle. However, through the co-ordination of the interaction design development of other countries, the development of China’s interactive design needs to be further comprehensive. Therefore, based on the research experience of this professional field, the application research of visual agile interactive animation design is the direction to explore the interaction design. Coordination of development trends and visual design interactions is essential.
Computer vision is a frontier of technology, widely used, and subject to important research, but computer vision is also a very high threshold technology. It requires developers to have a high theoretical foundation and practical hands-on ability, thus for beginners and non-technical sciences. It is difficult to learn and apply students. In response to this problem, researchers have developed some general systems and application libraries, such as OpenCV [1]. OpenCV is an open source computer vision library funded by Intel Corporation. It consists of a series of C functions and a small number of C++ classes, implementing many general algorithms for image processing and computer vision. OpenCV can also be used to perform advanced processing on images such as feature detection and tracking, motion analysis, target segmentation and recognition, and 3D reconstruction. OpenCV has lowered the threshold of application of computer vision to a certain extent. However, users must be very familiar with C/C++ programming, and spend a lot of time familiar with OpenCV API, which greatly increases the time period and complexity of system development. This article focuses on the field of computer vision and has developed a new type of interactive computer system. The system encapsulates the OpenCV library by visual programming, which minimizes the development and use threshold of the computer vision system. It uses the open source library QT to develop a beautiful and human software interface, and at the same time simplify user operations. This software is mainly for Windows platform users.
Animation based on animated visual effects is a system component
The structural system of this paper is shown in Fig. 1. It is divided into three layers from top to bottom: visual programming interface, various forms of modular system, and plug-in structure of the system.
Interactive system basic structure.
We use a node-based visual programming approach to achieve visualization of programming work, avoiding direct user interaction with code, and being more friendly than traditional code programming. The graphical user interface of the system in Fig. 2 is based on the C++ open source library QT development. The folder directory on the left side of the user interface is a list of nodes, and the gray area on the right is the workspace created by the user. The rounded rectangular object in the work area is the node referred to in this paper. The management of the node is based on QT’s graphics-view framework [2], and the node drawing adopts QT’s proxy window technology. The dots at the ends of the node represent the port, the input port is shown in blue, and the output port is shown in red. There are three kinds of nodes in the scene: only the data provider of the output port, such as the File Loader control that provides the file path; only the data consumer of the input port, such as the Image Displayer control that displays the image; both the input port and the output port. The data handler, the title bar of this node is green, such as the CvImage2QImage control in the converted image format. The workflow of user visual programming is: select the desired node type in the node list, create a node object in the work area; then use the mouse to hook the data stream between the ports with the same data type, specify the nodes between the data dependencies form a graphical flow chart. The nodes are the vertices of the graph and the data stream is the edge of the graph. The system automatically executes the scene graph created by the user and generates corresponding results to the user.
Data interaction diagrams.
This system has developed a powerful module toolbox. The modules in the toolbox encapsulate the commonly used algorithm functions in the OpenCV library, such as importing pictures, video modules, connecting camera modules, displaying images, video modules, and image, video analysis, and processing modules. Based on the off-the-shelf toolbox of the system, users can easily perform experiments such as image processing, video tracking, contour extraction, and face detection. The modules in the toolbox are divided into three types according to functions: input module, output module, and processing module. They correspond to the provider of the data in Section 2.1, the consumer of the data, and the processor of the data. 2 is a schematic diagram of data interaction between the system and the user. The interaction process is: the user passes the data to the data processing module through the input module, the processing module transmits the running result to the output module, and the output module visually feeds the result back to the user. The input and output modules support basic data types such as integers and floating-point numbers, as well as data types such as vectors, matrices, and even image files and video files.
System plug-in structure
The industrialization of software has led to the evolution of software reuse from general-purpose libraries to domain-oriented application frameworks. The reuse of application frameworks has become one of the most effective reuse methods in software development and production. In the face of this development trend, what is coming out is a new, open and highly scalable architecture system, namely plug-in architecture [3]. The benefit of the plug-in architecture is that it strips the extensions out of the framework, reducing the complexity of the main system framework and making the framework easier to implement. The extended functionality is coupled to the framework in a loosely coupled manner, and both can be independently changed and published while maintaining the same interface. A general-purpose computer vision system must be highly open, and it is convenient for users to develop specialized systems for application fields. Therefore, the system adopts a plug-in architecture system. The main program of the system is just an open application framework. The application framework provides a standard interface for plugins. The plugins exist in the form of DLL files. Users can call the open system package, develop their own input modules, output modules, processing modules, and then these functional modules are finally integrated into the system toolbox in the form of plug-ins. The user only needs to put the developed plug-in file into the system’s resource directory. When the main framework is running, the plug-in will be dynamically loaded into the system. It can be seen that the system has very high openness and scalability.
The embodiment of interactive animation vision
Touch feedback technology
For many new media art, touch is especially an important part of the touch screen is a device that makes multi-media information more intuitive and convenient. The new medium breaks through the traditional “only see, not touch”, more is to require the participation and cooperation of the audience, the information exchange between the sender and the recipient is two-way. There are many expressions of new media art, but there is only one commonality between them. That is, participants interact directly with the works. Users participate in and change the image, content and even meaning of the works. They trigger the transformation of the works in different ways: Touch, space move, sound, etc. For example, in the process of operation, the user can perform corresponding operations according to the prompts on the screen according to his own requirements. When the user touches the screen information, the screen on the screen will be displayed with the touch of the hand. Changes are made according to the programming of the program until the screen that people need is displayed.
Large screen presentation technology
The computer enables artists to create new works of art. These animated video works created under the new media conditions include three-dimensional digital images that are almost impossible to achieve by traditional manuals. The large-screen presentation technology has LED electronic display and display. Screen demonstration technology, projector demonstration technology, and the use of various projection light to illuminate animated images such as walls, water curtains, air, etc., and animated images placed on the screen can prompt more viewers to understand the story content of the story. And participate in the interaction process of interaction and experience, among which “2010 Shanghai World Expo City Earth Pavilion”, the exhibition mainly focuses on the relationship between human beings, cities and the earth is a win-win and symbiotic relationship. The blue planet in its pavilion, is a 32-meter-diameter “Earth” that will rotate in accordance with the direction of the Earth’s rotation. The surface uses large-scale multimedia seamless mosaic projection technology, using sound, lighting, indoor space and other elements to create a new real subject space. The design works simulate the aerial view of the Earth from the scene, thus allowing visitors Feel the fun of watching the Earth from space.
Virtual reality technology
“Virtual reality technology, which combines computer graphics, multimedia technology, computer network technology, simulation technology, artificial intelligence technology and multi-sensor technology to simulate human visual, auditory, tactile and other sensory functions, enabling people to immerse in In the virtual world generated by the computer, and by real-time interaction with the natural way of using language, gestures, etc., a humanized multi-dimensional information space is created” [4]. For example, the experience can be directly Touching the object simulated in the virtual environment, the hand has the feeling of holding something, and can feel the weight of the object, and the object caught in the field of view can immediately move with the movement of the experiencer’s hand, usually this virtual reality The technology enables the experience user to feel like an immersive experience and experience the fun of interaction. The use of virtual reality technology and the latest achievements of artificial intelligence technology are also applied in software development and new media art creation. more.
Animated visual effects interactive system example
Video motion capture
In this experiment, the camera captures the motion of one hand, and the size range of the shot is determined by image recognition. Then the image of the binary handle becomes a black and white image, and the image of the hand is composed of many pixels, and the coordinates of these pixels can be determined. The purpose of motion capture is to identify the motion parameters (slope) and position parameters (centre) of the hand, where the direction of the line is the direction of the palm (for clarity, the line is drawn outside the hand), and the white part of the palm is the centre of the palm. (Indicated by a short line), the coordinates of the palm centre are easy to calculate, and will not be described here. The slope of the palm is obtained by the least square’s method. The specific solution process is as follows:
A number of pixel point coordinates
Find the extreme value:
Finally get:
This will determine the equation for the line that replaces the palm.
It is assumed here that people stand and do not exercise, the movement is the arm, the hand movement captures the ball, then the ball movement, the hand and the ball need to do the collision detection. The arm movement uses the IK CCD method and the end effector is set to the palm centre. The motion solving techniques of joint chain structure mainly include forward kinematics and inverse kinematics. The so-called forward kinematics method is to determine the position of each joint and the position of the end effector by giving a state vector, and the inverse kinematics solves the state of the joint chain structure by giving the position X of the end effector. Vector, and then find the spatial position of each joint. The advantage of forward kinematics is that the solution process is intuitive and simple; the disadvantage is that there are too many input parameters, too many degrees of freedom, and the user is at a loss. The advantage of the inverse kinematics method is that the entire state vector can be reversed by simply giving the position of the end effector; the disadvantage is that there are multiple solutions and unstable solutions. For simple reverse motion (such as two links), you can directly find the analytical solution. For complex reverse motions (more than two links), the iterative method is usually used until the error between the end effector and the target is within the allowable range. This paper only considers the motion control of three joints. There are three chain rods, which can be solved by CCD method. The physical model of the hand reflects the actual slope of the hand in real time. The end influencer is the centre of the palm and is set in the middle position of the palm model.
Experimental results
The algorithm is implemented in Visua1C++, the graphics platform is OpenGL, and the experimental objects are hands and balls. First, image recognition determines the range of the hand, finds the centre of the hand, and determines the slope of the hand by least squares. Then, the parameters of the handle are passed to the physical shape, and a match is achieved between the two. The ball is shot by the movement of the hand, the ball is photographed, and the ball is moved, and the ball can be continuously taken. Figure 3 shows the system interface.
Hand shot animation.
The effect of the animation works and the evaluation system are established. Firstly, according to the target audience of the works and the psychological needs of the audience and the motivation of the audience, only the background of the audience of the creation of the works can be clarified, and the true standard of the works can be refined, which is based on the audience psychology and the audience. The accepted comprehensive standard is to prove that the final result of the work is successful. The audience’s reception and feedback are reflected in two aspects: First, the aesthetic psychology of anime audience and the picture of anime works, aesthetic psychology includes emotional aesthetics and visual aesthetics. Emotional aesthetics refers to the subject with aesthetic value to the subject. A beautiful cognitive tendency and special feelings, this is reflected in the expression of emotional themes in the content of the work. The second is “visual aesthetics, which means that the vivid, smooth, and colourful image gives the audience a visual sense of pleasure.” This refers to the formal beauty of the images in the visual communication design, that is, the use of certain modelling means and aesthetic symbols. Creation, so the beauty of the picture is very important to meet the aesthetic psychological needs of the audience.
The basic principles of the unity of content and form of successful cartoons are unchanged. Therefore, the standard of content and performance standards can be divided into two aspects, one is the evaluation criteria of the story plot, the second is the evaluation criteria of the design, and the formal beauty of the law as the main standard of visual communication design. It can be seen that only by understanding the internal needs and appreciation motives of the original animation of the audience can the effective acceptance of the animation audience be mastered, thus determining the aesthetic standard of the work and the survival value of the work. Therefore, to analyze the psychological needs of the animated audience and propose countermeasures for animation creation, we can create animation works that meet the internal needs of the audience.
Conclusion
The generation and development of interactive animation is closely related to the generation and development of computer, multimedia, network technology and interactive technology, and gradually formed and developed into an emerging digital animation expression. The art form of interactive animation is different from traditional animation. It is the combination of art and computer graphics. It is a new topic of high-speed development and combination of computer hardware and graphics algorithms. It uses computer science, art and other related disciplines to synthesize a variety of continuous virtual images on the computer, providing viewers with a new world to fully display their personal imagination and artistic talent.