GET IN TOUCH WITH PAKKO, CREATIVE DIRECTOR ALIGNED FOR THE FUTURE OF CREATIVITY.
PAKKO@PAKKO.ORG

LA | DUBAI | NY | CDMX

PLAY PC GAMES? ADD ME AS A FRIEND ON STEAM

 


Back to Top

NUCL3AR // DIGITAL ARTS

Computer Modeling of Visual Model of Animation Color Information Based on Augmented Reality Technology

Computer Modeling of Visual Model of Animation Color Information Based on Augmented Reality Technology

With the rapid development of computers, its application fields have also begun to spread to all walks of life, and the modeling of information visualization models for animation colors has also begun to spread. This paper was aimed at studying the computer modeling of the visual model of animation color information. To this end, this article proposes an optimization of the collection of animation colors based on augmented reality technology and image recognition technology to make the collected data more accurate. It provides accurate color information to the computer, making the color information of animation more visible. To this end, experiments and analysis are also designed to conduct a comprehensive analysis and performance exploration. The experimental results in this paper show that the improved visualization of the animation color information visualization model has increased by 27.4%, which effectively solves the current problems faced by computer modeling.

1. Introduction

The use of color is an important part of comic creation and plays an important role in the quality of comics. The use of color plays an important role in shaping the image of the character, describing the character of the character, and in expressing feelings and creating the atmosphere of the characters and scenes, promoting the development of the plot and enriching the content. The content and other aspects of the screen cannot be separated from the use of colors. As domestic animation has a unique style in the grasp and use of colors, the integration of traditional colors has promoted the uniqueness of China’s animation industry, has also achieved some achievements, and maintained a stage of continuous development. After liberation, everything in China was in a recovery period, and China’s animation industry also tended to sprout here. Starting from the development of modern Chinese animation colors, animation has gone through multiple growth periods in various stages. The color development of Chinese animation has traditional characteristics and the characteristics of the integration of Chinese and Western.

The production of Chinese animation is in modern times, and the grasp and use of animation colors are from scratch. It is constantly improving, constantly showing the light of Chinese traditional colors, and at the same time injecting the energy of foreign animation colors. Following the world’s high-level animation, in modern Chinese animation, the use of color has the characteristics of the era, content direction, cultural change, and the differentiation of the color performance of the animation. Therefore, through the study of the color characteristics of modern Chinese animation, the development method is combined with the use of today’s animation colors in China. Exploring China’s animation today is the most important part of the self-development process modeling, which is the importance of this article.

In the process of advancing the visualization of animation color information, the proposal of computer modeling has accelerated this process. At the same time, the development of enhanced display technology has also injected vitality into this industry, making more and more people begin to invest in related research. In their article, Ibanez et al. showed that augmented reality technology has a positive impact on the learning-related outcomes of Mexican middle school students. However, the impact varies depending on whether the student is a public school or a private school. They also designed an augmented reality application that allows students to practice the basic principles of geometry, and a similar application. It contains the same learning objectives and content and is deployed on the Web [1]. Barbin believes that the design of the nuclear reactor does not provide a solution for decommissioning, and there is no safety technology to deal with irradiated reactor graphite. The decommissioning of a uranium-graphite reactor is a complex combination of tasks involving the selection of appropriate methods and technologies for the treatment of radioactive graphite. The behavior of radioactive elements can be determined by computer modeling of the postprocessing of reactor graphite by heating in a carbon dioxide environment [2]. Petrov and Kiselev model the polarization characteristics of the random Gaussian particles that simulate dust particles on the surface of the weathered layer. Model calculations show similar behavioral patterns of polarized branch parameters. It studied the influence of the refractive index of a single scattering particle on the size and position of the maximum value of the positive branch of the linear polarization degree in the considered model [3]. The time dependence of the number of NBOHC calculated by Barannik et al. and the luminescence intensity of the red band associated with these defects are in good agreement with the experimental data. It allows estimation of the average cross-section generated by NBOHC radiation, the radiation-enhanced hydrogen diffusion coefficient, and the reaction rate constant of hydrogen injection to passivate NBOHC [4]. Mei et al. believe that sharing data for public use needs to be cleaned up to prevent leakage of sensitive information. Previous research has proposed ways to create privacy protection visualizations. However, few of them provide users with sufficient feedback on how much utility is reduced (or retained) in the process. In order to solve this problem, they designed a visual interface and a data operation platform [5]. Xiao-Jie et al. analyzed 3,060 articles published in Advances in Psychological Science from 1983 to 2014, using the information visualization method of the CiteSpace software to analyze the number of publications and the frequency of citations, etc. In addition, it also analyzes downloads, funding, organization, author, and keywords. The results show that the amount of literature published every year is on the rise, and 49.4% of the papers are supported by national or provincial-level projects [6]. In their research, Akle et al. proposed a response quality index to calculate the designer’s selection quality for any given scene and a total of 30 participants took 90 tests. The parallel coordinate chart proved to be the best chart chosen in DesignbyShopping [7]. Byrne et al. critically analyzed the scope and applicability of different visualization techniques for information visualization practitioners. They believe that graphic elements are widely used in information visualization in practice and are increasingly regarded as conducive to memory. In order to support the critical analysis of hybrid visualization, they provide a visualized image information content model, called the visualized frame model [8]. The above several articles are quite in place for the related information visualization technology and the description of the enhanced display technology, and there are detailed instructions for the use of key technologies and the problems in the use process. However, the explanations are basically centered on a key point, and the two are not combined to conduct a systematic study.

The innovation of this article is to use augmented reality technology, image recognition algorithms, and animation color classification as theoretical support to optimize the existing visual computer modeling of animation color information. It optimizes the image recognition algorithm to make the information collection of animation color more accurate. And through the experimental part of two unrelated data sets for an animation color collection to test, compare the improvement of its collection ability.

2. Augmented Reality Method

2.1. Augmented Reality Technology

Augmented reality is an important field of virtual reality technology. The ultimate goal of augmented reality technology is to seamlessly embed computer-generated virtual information into the real world that users can observe. The use of augmented reality technology can extend the user’s visual function, improve the user’s visual experience, and dialogue with the environment where the real world and virtual information overlap [9].

The use of computer technology to embed virtual information in the real scene is also known as the extended reality of compound reality [10]. Real-time overlapping of real scenes and virtual information in the same screen or space achieve the harmonious coexistence of the virtual world [11]. The continuum of virtual and real space is shown in Figure 1:

The continuum of virtual and real spaces
2.1.1. Principles of Augmented Reality Technology

In order to achieve the success of the augmented reality technology system, it is very important to accurately locate the virtual object on the actual object and seamlessly integrate the actual scene of the surrounding world with the virtual extended information generated by the computer [12]. The method of setting relevant information (manual marking, extract marking information, and obtaining information required for registration) in the actual environment and realizing three-dimensional registration is general. As shown in Figure 2, the complete augmented reality system consists of many parts (including markers, computers, cameras, display devices, and augmented reality software systems). Among them, the software system is a very important part of the augmented reality technology process [13].

The working principle of augmented reality technology
2.1.2. Implementation of Augmented Reality Technology

There are many ways to implement enhanced technology. At present, the general opinion in the academia is that the implementation of augmented reality technology is a network terminal.

With the development of computer and communication technology, networked augmented reality technology is an important development direction [14]. Network applications can effectively reduce equipment costs and expand system application space. Many years ago, the concept of a networked augmented reality system was proposed. The augmented reality system can be connected to the Internet through GPS and wireless access to the network (see Figure 3). Now, this idea has become a reality. Tentacle Technology has developed an augmented reality Web program based on Flash technology. This does not require software installation and plug-in downloads but provides services to Web users based on augmented reality [15]. Through the combination of Web-based augmented reality technology and network technology, a very effective augmented reality interactive experience can be achieved (mainly through the realization of remote teaching, simulation experiments, extended technology games, product displays, etc.) [16].

The Internet through GPS and wireless access to the network.
2.1.3. Tracking Location Registration Technology

This technology mainly locates and recognizes images or targets in reality and converts these positioning targets into computer virtual information for storage through computer technology [17]. According to the different positioning technology, we mainly divide it into location-based and image-based systems. The latter can be divided into marked type and unmarked type according to the actual situation. For the position tracking system, it mainly uses a large number of magnetic tracking and global positioning system to achieve precise positioning and obtains the corresponding position data through precise positioning and then transmits it to the user [18]. For example, our common EnvironmentalDetectives software uses GPS to track and locate. For the marker-type image positioning system, it mainly locates the target through a large number of manual markers. This calculation method and the ground reduce the calculation complexity and the calculation efficiency is higher. The unmarked system mainly presents information based on the characteristics of real objects. For example, the BMW car maintenance system is to quickly locate and recognize the real parts of the car. Through the analysis and summary of a large number of research materials and practical cases, we found that the currently widely used technology is the positioning and tracking technology based on image recognition [19].

2.1.4. Display Technology

This technology mainly presents the virtual information stored in the computer in the previous link and forms an image that simulates the real scene after a certain amount of data is transformed. At the beginning of the research, this kind of reality technology is mainly based on helmet-mounted displays, mainly including optical see-through and video see-through. The former is mainly based on semitransparent and semireflective optical equipment for simulated imaging. This design allows the experimenter to observe the limited space entity and the virtual space processed by the computer at the same time. The latter mainly uses a camera as a tool to shoot objects and displays the real scene through the fusion of shooting information and computer virtual information [20]. In contrast, the information processed by the light see-through display is relatively simple. It only needs to process the virtual information of the computer, and the video see-through display needs to process the information generated by the camera and the computer at the same time [21]. At present, the display devices that people are more exposed to are mainly computers, tablet computers, and other devices, while helmet-mounted displays are less used in people’s real life. It is mainly aimed at the more professional aerospace and other fields. With the continuous development of science and technology, the development of smart mobile devices is very rapid, and the size of the display gradually tends to be miniaturized. The size is basically above 3.5 inches, and the resolution has also been greatly improved [22]. Compared with previous devices, display devices are developing in the direction of volume, lightweight, and strong operability, and their development prospects are very broad.

2.1.5. Application of Augmented Reality Technology

No matter how precise the drawings are, it will also limit the accurate expression of the designer’s design concepts and affect the communication with customers. The augmented reality technology just makes up for this shortcoming [23]. It integrates image recognition, motion capture, and virtual reality to accurately superimpose digital information and three-dimensional virtual models into the real scene with innovative human-computer interaction technology. The specific application is shown in Figure 4:

Application of augmented reality technology.

Augmented reality technology can not only be applied to the above-mentioned fields (exhibition, marketing, science and education, design, publishing, entertainment, and other fields) but also can perform remote real-time collaboration through simulation experiments and continuously produce new research results with low cost and high efficiency.

2.2. Image Color Recognition Based on Support Vector Machine

Compared with previous general mechanical learning methods, support vector machines have excellent general generalization capabilities, and their complexity has nothing to do with the dimensionality of the sample. The structure of the support vector machine is similar to the three-layer feedforward neural network, and the number of hidden layer nodes is determined by the support vector [24, 25]. By solving the optimization problem of convex secondary programming, the number of hidden layer nodes and the weight vector can be obtained at the same time.

Assuming there are training samples (, ), where , the hyperplane is shown in Equation (1):

The sample must meet the following conditions:

It can be seen from the above formula that the distance between the support vector and the classification plane is the closest, and it is on H1 and H2. And the solution of the above function is obtained by constructing the Lagrangian function:

Among them, is the Lagrangian multiplier.

Let the partial derivatives of and be equal to 0, then,

Substituting it into Equation (4) can get

According to the wolf duality theory, the dual problem of the optimization problem is obtained:

According to the solved , we get

The decision function is

For the case where the sample is not linearly separable, a nonnegative slack variable is introduced:

Construct the optimal hyperplane and classify the sample with the smallest error:

Among them, is the penalty factor, which represents the degree of punishment for errors. The larger the , the greater the penalty for misclassification.

Among them, 1 is the Lagrange multiplier.

In the same way, the secondary planning problem can be obtained:

For nonlinear problems, by adopting the appropriate kernel function , the nonlinear transformation of the input space is mapped to a linear problem in a high-dimensional space. The above situation is transformed into

According to the wolf duality theory, the dual problem of the optimization problem is

The decision function is

Commonly used kernel functions have radial basis kernel functions, because the feature space corresponding to the radial basis kernel function is infinite. In general, a limited sample must be linearly separable in the feature space, so the radial basis kernel function is the most commonly used kernel function:

The optimization of image recognition can improve the accuracy of obtaining animation color information, and the ability to obtain color has been improved.

2.3. Animation Color Classification

The expression of animated films is more subjective, more flexible, and more free to play than real movies. Especially in terms of color, many colors in animation are impossible to see in real life. Whether it is the performance of light in the animation or the reproduction of natural scenery, or the animation director has added his own subjective creation of color for the needs of the animation plot, these have greatly enriched the virtual scene screen of the animation film, making color animation even more eye-catching.

2.3.1. Light and Shadow Color

Through the study of color composition, we know that the essence of color is light. Where there is light, there will be shadows, and vice versa. Therefore, light and shadow are indispensable in artistic performance, especially in film and television. “Light and shadow can not only express the volume, spatial relationship and texture of the subject and other image modeling factors, etc. From the perspective of cartoons, the traditional two-dimensional animation pursues the artistic charm of simplicity and generalization in the use of light due to technical limitations or the pursuit of stylization.” Therefore, in the animation film, the use of light and shadow should be based not only on the style of the film but also the color of the light and shadow according to the needs of the picture. However, due to the development of current three-dimensional technology, the development of three-dimensional animation films has gradually grown. The development of software technology has made the use of light and shadow in animation films more delicate and natural, and more realistic and accurate in scene performance and character modeling. Since each cartoon has its own style, there are different ways of expression in the use of light and shadow. Figure 5 shows the three primary color models.

Color three primary color model
2.3.2. Natural Sketch Color

The colors in nature are rich and colorful, and they are the foundation and source of art. So many artists draw inspiration from nature when they create artworks. Nature gives life a variety of beauty, especially in terms of color. The color creation in animation films is also inseparable from the influence of colors in nature (Natural color is a record of life and a guarantee for the vitality of animation works. The painterly sketch color is the foundation and source of the animation color, and the animation color is the development and change of the natural sketch color, it is the rational generalization and creation of the natural color.), so in the process of animation creation, we will find some colors in nature. As the basis of creation, subjective processing and recreation of the colors can be based on doing some conscious artistic effect processing, showing it through a series of elements such as the lens scene. Especially in animation scenes, the realism of color is the most basic and most important.

2.3.3. Subjective Creative Colors

If film and television art is a kind of expressive art of light and shadow, then color is an indispensable element of expression in film and television art. The accuracy of the use of color is directly related to the expression of audiovisual language in film and television works. A large part of the expression of color in film and television works uses subjective color with a certain metaphor or symbolic meaning, especially the color expression in animated films. And animation will as a new force in film. The hypothetical exaggeration of the animation itself determines that it is better than the actual film in the subjective expression of color. Subjective animation colors are very subjective and creative in expressing the modeling of animated characters and portraying the character’s psychology, which dyes the atmosphere of the picture and embodies the emotions. It uses more lively colors, is more bold and exaggerated, and can be free from the constraints of objective colors. The television is shown in Figure 6 as a specific color analysis model.

Color analysis model
2.3.4. Uniqueness of Animation Color

Color is used as a symbolic element in film and television works or in painting art, showing different pictures. Film and television animation and real movies belong to the same category of film and television art. Whether it is a film animation or a real movie, although there are similar techniques in the expression of colors, due to the characteristics of animation art, this determines that the animation color has a richer and more unique nature.

2.3.5. The Creativity of Animation Colors

Although the color of animation is a product of a combination of subjective and objective, animation is a highly concentrated expressive art that conveys different feelings to the audience. Especially in the use of the color of the animation film picture, it needs a multidimensional expression technique. Animation colors not only meet the basic needs of the audience but also creatively show different colors.

2.3.6. The Hypothetical Nature of Animation Colors

The important feature of animation in realizing film creation and performance skills is extremely hypothetical, and its main feature does not pursue the realism of the picture. In the process of creating vision, the creation process of animation artists is different hypothetical assumptions in various aspects. Therefore, this hypothetical nature of animation art determines that the color of animation is also a hypothetical color. The assumptions of animation colors are expressed in the animation character modeling, the assumption of the scene, and the assumption of the form of the picture.

2.3.7. The Ideographic Nature of Animation Colors

The subjective, virtual, and creative characteristics of animation art determine that animation colors are different from ordinary colors. But for the ideographic nature of color, this means that the color in film and television animation respects the expression of ordinary film and television color language in the expression of color language, and is different from ordinary film and television color language.

2.3.8. The Associativity of Animation Color

Color itself has a certain association, for example, from red to flame and red flag. And the colors in the cartoons also continue this associative nature of colors. However, film animation is different from ordinary painting, so the color association of animation is not only the association of static colors, it can also be the color association of dynamic pictures or the color association of dynamic music.

3. Experimental Results and Analysis

3.1. Animation Color Collection Experiment

In order to explore the problem of the accuracy of the improved image collection for the collection of animation colors, this article designs related experiments to simulate the collection process. The experiment in this section uses color collection of 1000 types of animated images on the Internet and divides them into two different samples. The number of each sample is 500, and the color collection software before the improvement and the animation color collection after the improvement are used to collect the color of the animated images of the two sample sets.

For the case where the training data is an image, due to the strong correlation between adjacent pixels, the input data during training is redundant. To be more precise, we can make the input of the algorithm (The purpose of whitening is to reduce the redundancy of the input through a certain transformation) so that the input of the algorithm has such properties:
(1)The data has low correlation(2)All features have the same variance. Commonly used whitening methods are PCA whitening and ZCA whitening

Use the principal component analysis method to divide the experimental samples, and the results of the division are shown in Table 1:


Color type Sample set 1 Sample set 2

0-20 87 65
21-40 92 88
41-60 114 146
61-80 129 74
81-100 78 127

Division of experimental sample set.

First, we use the color collection of the animated image before the improvement to collect and train the two samples and count the accurate number of his collection, accuracy rate, and other parameters. The collection results of sample set 1 are shown in Table 2:


Color type Sample set 1 Correct number Number of errors

0-20 87 62 25
21-40 92 71 21
41-60 114 84 30
61-80 129 94 35
81-100 78 52 26

The collection results of sample set 1 by the improved collection method.

At the same time, the second sample is collected and trained using the color collection of the animation image before the improvement, and statistics of his collection accuracy, accuracy rate, and other parameters are counted. The collection results for sample set 2 are shown in Table 3:


Color type Sample set 2 Correct number Number of errors

0-20 65 42 23
21-40 88 61 27
41-60 146 101 45
61-80 74 42 32
81-100 127 89 38

The collection results of sample set 2 by the improved collection method.

Then, use the improved animated image color collection to collect and train the first sample, count the number of collections and accuracy rates obtained during the collection process, and draw it into Table 4:


Color type Sample set 1 Correct number Number of errors

0-20 87 81 6
21-40 92 82 10
41-60 114 98 16
61-80 129 106 23
81-100 78 69 9

The collection results of sample set 1 by the improved color collection method.

At the same time, use the improved animation image color collection to collect and train the second sample and count the accurate number of his collection, accuracy rate, and other parameters. The collection results for sample set 2 are shown in Table 5:


Color type Sample set 2 Correct number Number of errors

0-20 65 61 4
21-40 88 83 5
41-60 146 131 15
61-80 74 68 6
81-100 127 117 10

The collection results of sample set 2 by the improved color collection method.

By comparing the above table, it is not difficult to see that the accurate number of acquisitions before the improvement is basically maintained at the level of one and a half. After improvement, the accurate number of information collected for animation colors can basically be called good, and there is still a very good improvement.

3.2. Visual Analysis of Animation Color Information

The choice of color is self-evident for the importance of information visualization. Color can not only distinguish the relationship between various events and data and help users remember information but also express the emotional attributes and atmosphere of abstract information.
(1)Enhance the contrast of types and colors. When dealing with the color of interactive information visualization in Web pages, the corresponding relationship between graphic size and color is a key design factor. As shown in Figure 7, there are 5 colors listed in the left image. If they are arranged side by side in the form of squares, it is easy for users to distinguish; when it appears in the form of a line chart, the difficulty of distinguishing colors is significantly increased, especially blue-green and yellow-green; when it is further used to represent a scatter plot, the color is even more difficult to identify

Color comparison diagram.

The data research of the survey proves that when the information in the Web page is visualized as the color choice, the appropriate increase of the color hue type and the change of the brightness contrast can make the color easier to distinguish and identify. A strong color such as big red, when used together with weaker colors or black and white, can play a prominent role and show its importance. Therefore, after resetting the colors in the right picture, the five-color line chart in the circuit diagram is more obvious than the left picture. The dark blue dots in the scatter diagram are the most obvious, and light blue is easier to identify, while the contrast between orange and pink, and green and blue is difficult to distinguish. It can be concluded that when visually designing colors, in order to distinguish smaller graphic elements, it should be set to a color with a larger contrast with the background, for example, dark blue graphics on a white background and yellow graphics on a black background, to increase the user’s easy recognition.
(2)Color analysis and comparison. At the same time, this article uses image recognition to visualize the animation color information, and comprehensively analyze and model the obtained results. The analysis results of the animation color information collection before and after the improvement are shown in Figure 8:

Comparison of color collection and analysis before and after improvement.

The “model for modeling” framework has been widely used in science teaching in the past ten years to guide and design model-based science teaching activities. Research has shown that these teaching activities help students engage in active and meaningful learning, especially to improve the understanding of the three-level representation and the ability of mental transformation between the three-level representation, so as to realize a reasonable understanding of complex chemical thoughts and knowledge, and enhance the metavisualization ability. The main reason why this mental processing process can be carried out smoothly and effectively is the full experience and exciting modeling links, as well as the explanations and predictions made by the modeling. Modify the model originally designed by analyzing different situations to adapt to different needs. For this reason, the similarity between the model and the actual effect in the modeling process must be guaranteed. For this reason, this article makes a comprehensive comparison of the modeling effects before and after the improvement. The modeling of the two data sets before the improvement is shown in Figure 9.

Figure 9 
Modeling similarity of the two data sets before improvement.

By comparing the modeling similarity of the two data sets before the improvement, we can see that the similarity of that is as follows:
(i)For the 0-20 data set 1 before the improvement is 71.26%. For 21-40 categories of data set 1, the similarity is 77.17%. For 41-60 categories of data set 1, the similarity is 73.68%. For the 61-80 category data set 1, the similarity is 72.87%. For data set 1 of 81-100 categories, the similarity is 66.67%.(ii)Before the improvement, the similarity of data set 2 of the 0-20 category was 64.62%. For 21-40 categories of data set 2, the similarity is 69.32%. For 41-60 categories of data set 2, similarity is 69.18%. For the 61-80 category data set 2, the similarity is 56.76%. For the data set 2 of 81-100 categories, the similarity is 70.08%.

Generally speaking, the level is not enough, at a poor level, and the degree of modeling and simulation is not high. At the same time, the improved modeling method is used to model two different data sets, as shown in Figure 10.

The improved modeling similarity of the two data sets.

By comparing the above picture, we can see that
(i)The improved method has a similarity of 93.1% for data set 1 of 0-20 categories. For 21-40 categories of data set 1, the similarity is 89.13%. For 41-60 categories of data set 1, the similarity is 85.96%. For 61-80 data set 1, the similarity is 82.17%. For the data set 1 of 81-100 categories, the similarity is 88.46%.(ii)Before the improvement, the similarity of data set 2 of the 0-20 category was 93.85%. For 21-40 categories of data set 2, the similarity is 93.32%. For 41-60 categories of data set 2, the similarity is 89.73%. For the 61-80 category data set 2, the similarity is 91.89%. For the data set 2 of 81-100 categories, the similarity is 92.13%.

It can be seen that the visual modeling similarity of the improved animation color information is basically maintained at about 90%, which is significantly better than the modeling similarity before the improvement.

Based on the above analysis, it can be obtained that the improved visual modeling of animation color information has improved color contrast, color matching, and the accuracy of collection and modeling similarity to varying degrees. Among them, the modeling similarity is increased by 27.4%, which can effectively improve the problems in the modeling process.

4. Conclusion

This article mainly studies the construction of computer modeling for the visualization of animation color information, through the use of augmented reality technology, combined with image recognition algorithms to accurately collect the colors of the animation and provide the data to the computer for modeling. At the same time, in order to understand the color of animation, a considerable degree of research and analysis has been carried out on the classification of animation colors, and designed an experiment to explore the color recognition of animation. Finally, in the analysis part, for the accurate recognition, color matching and analysis, an exploration has been carried out on the modeling effects of two different data sets to ensure the feasibility of this method.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare no conflicts of interest.

This content was originally published here.