Strategies and Practical Challenges of Integrating Information Technology in Traditional Art Education

From the perspective of educational psychology, authentic and vivid scenario teaching is conducive to the active construction of students’ knowledge. The real immersion of virtual teaching eliminates the boundaries between students and computers and other equipment. Students as the protagonists in the virtual space, an all-round, immersive experience, so that it is difficult for students to realize that they are in the virtual environment, to create a kind of pass-through true experience. Google has developed a “Google Art Project” that allows students in the classroom to virtually view works of art from museums around the world that Google has captured with its ultra-high-resolution cameras. These works can be zoomed in on, allowing students to see them up close, deconstruct them in detail, and study them in depth. Students can directly feel the three-dimensional art works in the virtual environment, and the presentation of art works is not just pictures or videos, but from two-dimensional to three-dimensional stereoscopic, which, to a certain extent, changes the way students appreciate art works, which is difficult to do in the traditional teaching methods.

Participatory and interactive teaching content can effectively stimulate students’ interest in learning, and have a positive guiding effect on the construction of students’ new knowledge. The realistic sense of presence and immersion of virtual reality technology, like real life, students can interact with objects in the virtual environment, and can touch and rotate the objects in the virtual environment with their hands. For example, domestic and foreign classic paintings, architecture, ancient pottery, domestic and foreign sculptures, which cannot be displayed live in the regular classroom, are reproduced by using virtual reality technology, so that students can observe and feel them at close range without leaving home, so that students can have different learning experiences of the art works, thus enhancing students’ student interest. With the development of virtual reality technology, virtual teaching is also being used by teachers in secondary school art and drawing classes. In the teaching of sketching structure, students can observe the perspective structure of objects inside and outside from multiple angles; in the study of light and shadow, by switching the light source from different angles, students can observe the subtle changes of light and darkness of the objects; in the study of environmental color, students can observe the influence of the environment on the objects in detail. The unique advantages of virtual reality technology in traditional art education can enable students to interact more with art works and enhance their learning experience.

Using virtual reality technology, it is possible to simulate environments that exist in reality and virtualize environments that do not exist objectively. Such as students and virtual artists, communication and discussion through virtual technology, sharing the creative experience of virtual artists, and so on. The virtual explanation adds humane color and fun to teaching, and for students in rural areas, it enables resource sharing and educational equity. Virtual reality technology combines visual, auditory and tactile senses, constantly innovates and explores, and puts a variety of elements in front of the students in an all-round and multi-angle way, stimulates the students’ perspectives and tentacles, expands their cognitive scope and thinking space, enhances their dispersive thinking ability, and fosters their creative consciousness and creative inspiration.

The global illumination technique is primarily designed to simulate light that strikes a point and is reflected, or reflected many times, by objects other than from the direct light source. Incident light L_i item represents the specific value of celestial light such as sunlight and artificial light such as man-made lamps. The f_r term represents the actual light interacting with the object, reflecting, refracting and other phenomena, taking into account the physical properties of the different materials on the surface of the object on the impact of the light situation.

Embodied in the rendering equation, that is, the incident term L_i(p, ω_i) contains not only the radiant brightness radiated from the light source, but also other points to be rendered out of the radiant brightness. So the rendering equation is recursive.

The right part of the rendering equation, the bi-directional reflection distribution function BRDF in the self-luminous and reflective terms is determined by the object’s own properties and is a known term given when setting up the objects in the virtual scene.

The remaining part is the integral equation (1)∫Ω+Li(p, ωi) (n⋅ωi) dωi $$\mathop \smallint \limits_{{{\rm{\Omega }}^ + }} {L_i}(p,\>{\omega _i})\>(n \cdot {\omega _i})\>d{\omega _i}$$

In the calculation, the angle between a specific incident direction and the surface normal vector can be found by calculation, and then the result can be found by making a spherical integral for each direction. That is, the relevant terms of the integral equation can be determined by calculation.

Therefore, the right-hand side of the rendering equation has only L_i(p, ω_i) term, which is unknown due to the inclusion of the contribution of the terms of the outgoing results from other sibling rendering points.

At this point, the radiant luminance is set to L, the self-luminous term to E, and the known part of the reflection term as a whole is noted as X.

Then the rendering equation can be abbreviated as (2)L=E+XL

After a simple term shift and deformation, we get (3)L=(I−X)−1E

Where (I − X)⁻¹ is similar to 11−x and can be expanded as a power series when x ∈ (0, 1) is satisfied. And the reflection term in passing through different objects, there will be part of the energy is absorbed, resulting in energy loss, so the total energy can only be reduced to meet this condition. Therefore, it can be expanded as (4)L=(I+X+X2+X3+...)⋅E

That is, the radiant luminance (Radiance) to be solved for can be expressed as (5)L=E+XE+X2E+X3E+...

Where E is the direct illumination from the light source and XⁿE is the result of reflecting the light n times. The result of one reflection XE is called direct illumination in the rendering domain, the result of two reflections X²E is indirect illumination, and the sum of all the terms in the right part of the equation is called global illumination.

Ray tracing is one of the better ways to achieve a global illumination effect. In order for the result to more closely resemble the effect of natural light bouncing around the environment thousands of times, ray tracing needs to start with a light source, and keep counting the light hitting an object for a primary reflection, and the resulting secondary reflections of the reflected light hitting other objects (or the rest of the object) ...... and so on, until the final rendered The final rendering does not stop until the result reaches a certain level of clarity. And this is just the process of tracking a beam of light. By sampling the light source many times and performing this sequence of calculations on all the sampled rays, we can finalize the simulation of the whole scene needed in art education, allowing students to learn in a virtual scene.

In order to have a more intuitive form of light understanding in real-time rendering, i.e., to consider whether incident light in a certain direction is visible at the currently shaded pixel point, a visibility item is added to the calculation. (6)Lo(p,ωo)=∫ω+ Li(p,ωi) fr(p,ωi,ωo) cosθi V(p,ωi) dωi $${L_o}(p,{\omega _o}) = \mathop \smallint \limits_{{\omega ^ + }} \>{L_i}(p,{\omega _i})\>{f_r}(p,{\omega _i},{\omega _o})\>\cos {\theta _i}\>V(p,{\omega _i})\>d{\omega _i}$$

where L_o(p, ω_o) is the outgoing light, L_i(p, ω_i) is the incident light, f_r(p, ω_i, ω_o) is the BRDF with clamping weights, and V(p, ω_i) represents the visibility.

The ambient light terms added from the reflection equations to the rendering equations and in the classical Blinn-Phong lighting model are approximated as constants. In screen space, since the raw geometric information of the scene has already been processed, leaving only the depth, normals, etc. recorded in two dimensions, and since the overhead of many methods lies in the fact that the process of finding the incident light leads to a large number of additional computations, it is useful to assume constant indirect illumination, modeled after constant ambient light, i.e., letting L_i be a certain value.

Furthermore, consider the nature of the integral. Usually the integral of a product is not equal to the product of integrals, but an approximation exists. Consider the L_i and BRDF terms of the rendering equation. When the BRDF is highlight, its outgoing direction region is small in extent, i.e., the contribution of this incident light to the coloring result is small. When the BRDF is diffuse, it reflects in all directions, but the transition between different directions is smoother. In other words, when the BRDF is highlight or diffuse reflection, one of the conditions under which Eq. (7) is approximately valid is satisfied, respectively. (7)∫Ωf(x)g(x)dx≈∫ΩGf(x) dx∫ΩGdx⋅∫Ω g(x) dx $$\mathop \smallint \limits_{\rm{\Omega }} f(x)g(x)dx \approx {{\mathop \smallint \limits_{{{\rm{\Omega }}_G}} f(x)\>dx} \over {\mathop \smallint \limits_{{{\rm{\Omega }}_G}} dx}} \cdot \mathop \smallint \limits_{\rm{\Omega }} \>g(x)\>dx$$

This method focuses on the case where the BRDF is a diffuse reflection-like material, which satisfies the conditions for the establishment of this approximation.

At this point, the L_i term and the BRDF term in Eq. (6) can be raised from within the integral, and the BRDF of the diffuse reflection material is known to be constant ρπ , thus simplifying the rendering equation to (8)Lo≈ρπ⋅Li⋅∫Ω+V cosθi dωi ${L_o} \approx {\rho \over \pi } \cdot {L_i} \cdot \mathop \smallint \limits_{{{\rm{\Omega }}^ + }} V\>\cos {\theta _i}\>d{\omega _i}$

The remaining variable to be calculated then is the visibility V term of whether or not one can be affected by indirect light from this constant at different coloring points.

After the analysis of 2.2.2, what this method does for the calculation of indirect light is mainly to find out the integral part of the simplified equation (8) for the visibility, i.e., the visibility of the hemispherical region on the surface of the current coloring point in each direction, so as to decide whether the indirect light irradiated to the current point in different directions can contribute to the current point. Finally, the total amount of indirect light, which is assumed to be a constant value, is subtracted from the sum of all non-contributing indirect light directions to obtain the portion of indirect light energy that can affect the current point, thus simulating the effect of global illumination.

The following three steps need to be completed in the main coloring calculation process:

In the first step, points are randomly sampled within the surface hemisphere centered on the currently computed coloring point. This step corresponds to the integration of the visible term V in the simplified Eq. (8), i.e., by summing up the results of the sampled points as the integral value. The sampled “surface hemisphere” is the hemispherical range of the color point towards the environment, i.e., the hemispherical range of the part that may receive indirect light. This range needs to be obtained by constructing the tangent space according to the normal information of the screen space scene obtained after the geometric processing stage.

In the second step, it is determined whether the sampling point has an occlusion effect on the current coloring point in the scene. This step makes a judgment by using the input scene depth information data saved in the screen space. If the depth of the sample point is smaller than the depth buffer record value (relative to the camera depth), it means that the sample point blocks the indirect light from the direction of the sample point that can be obtained from the region of the current calculation of the coloring point. Then the coloring point location should show a reduction in indirect light energy.

In the third step, the sampling results are analyzed. After random sampling and all the judgments are finished, the sum of the results of each sampling point is normalized with the total sample size, and the value is used as the attenuation of the indirect light received by the current coloring point, so as to simulate the phenomenon of indirect light being blocked in part of the direction.

After the application of virtual reality scenes for art teaching, it is necessary to analyze the students’ interaction experience, and the way of analysis is mainly student interviews and research, statistics on students’ views and evaluations of the virtual reality environment, virtual interaction behaviors, and virtual reality interaction, focusing on the analysis of students’ attitudes towards the interaction behaviors of the virtual reality scenes, with a special focus on the amount of effort students need to spend in order to complete the interaction task.

Through the research on students’ scene experience as shown in Figure 1, 45.79% and 26.02% of the students thought that different virtual reality scenes had a large or certain impact on students’ behavior of engaging in learning, respectively, while about 11.38% of the students thought that the impact was great, and only 16.81% thought that the impact was very small or almost no impact. Through this set of data, it can be concluded that different virtual reality scenarios have a large impact on the interaction between students and art works, including a variety of physical and psychological effects, so teachers should fully consider the production scenarios and appreciation contexts of art works when carrying out the interaction design of art works, and explore the different processes of virtual reality interaction design for different scenarios.

Figure 1.

Influence of different virtual scenes on students’ interaction with art works

Another important index to analyze students’ interaction experience in virtual reality scene is the degree of students’ completion of the set interaction tasks, which can reflect students’ interaction experience from the side by analyzing whether the students can complete the interaction tasks smoothly and whether other obstacles have appeared in the interaction process. By investigating the smoothness of the students’ interaction experience, as shown in Figure 2, 84.56% of the students can complete the simple interaction tasks normally or smoothly without being affected in the simple interaction scenario, which indicates that it is relatively feasible to carry out the interaction design research in the virtual reality scenario and the virtual scenario has a high simulation degree of the real thing, so the students’ interaction experience is smooth.

Figure 2.

Influence of different virtual scenarios on students’ interactive task

After conducting the analysis of student experience, the research method is summarized. Based on the special characteristics of virtual reality scene, the research method of interaction design in virtual reality scene is different from the traditional interaction design research method, and based on the students’ evaluation of virtual reality interaction, it can be concluded that the students’ views on the interaction design process of artworks in virtual reality scene under the students’ perspective. By investigating the students’ views on the interaction experience of virtual reality scenes as shown in Figure 3, it is concluded that 56.13% of the students believe that the interaction design methods in virtual scenes have a positive impact on enhancing the interaction experience of students, 23.69% of the students are neutral and believe that there is some impact, and the remaining 20.45% of the students believe that the impact is very small. Through analysis, it is concluded that the interaction design research method of artworks based on virtual scenes simplifies the traditional interaction design research process by combining virtual and reality, utilizes the high immersion of virtual reality scenes, and allows students to participate in the interaction design research process, which has a positive effect on enhancing students’ experience.

Figure 3.

The impact of interaction design methods on students

The data of the first course-related measurements are shown in Tables 1 and 2, from the mean value, the state of heart flow as well as the quiz scores of the experimental group are slightly higher than those of the control group, indicating that the virtual reality technology has an effect on the two factors; the learning engagement degree of the two groups is not much different. And from the variance, the two classes have similar magnitude of fluctuation in the scores of the four dimensions, indicating that the effects produced by virtual reality technology support and traditional classroom teaching will be affected by individual differences.

The second course-related measurements are shown in Tables 3 and 4, due to the original course content, no paper quiz questions were set this time, so there is no learning achievement score, from the mean value, the control group of traditional teaching is lower than the experimental group supported by virtual reality technology in in creativity, from the state of the heart flow as well as quiz scores, the experimental group is slightly higher than the control group, which indicates that virtual reality has an impact on the two factors; the learning engagement degree of the two There is not much difference between the two. In terms of variance, the scores of the two classes in the four dimensions show similar fluctuations, which indicates that the effects of virtual reality technology-supported and traditional classroom teaching are affected by individual differences.

Descriptive statistical analysis showed that the overall scores of the experimental class were higher than the control class only by means. Because the questionnaires completed by the two classes were independent of each other, an independent samples t-test was used to nearly determine if there was a significant difference between the two classes, the experimental class and the control class, after the use of virtual reality technology.

Original hypothesis: H₀: There is no significant difference between the two classes on the four dimensions.

From Table 5, it can be seen that Levine’s test of equivalence of variances significance are all much greater than 0.05, so we should look at the assumption of equal variance row, as can be seen from the table, the four dimensions are creativity Sig=0.002, mental flow experience=0.024, quiz scores=0.025, learning engagement=0.000, are all much less than 0.05, so the original hypothesis should be rejected and it should be considered that the two classes of the experimental class and the control class have no significant differences on the Creativity, Heart Stream Score, Learning Achievement Score, Learning Engagement Score are all significantly different, and the impact of using virtual reality technology on teaching and learning is very obvious. It is noted here that the significance of the heart flow score is 0.024 which is relatively large, analyzing the reason behind this may be that the class time is relatively short at 40 minutes, so the students do not have enough immersive experience, but it can still be seen that the use of a small period of time can make the two classes have a significant difference in the heart flow score, which indicates that virtual reality technology has a positive impact.