Research on the Implementation Path of Teaching Innovation of Higher Vocational Cultural and Creative Professional Group Empowered by aigc Technology
Published Online: Mar 24, 2025
Received: Nov 01, 2024
Accepted: Feb 13, 2025
DOI: https://doi.org/10.2478/amns-2025-0741
Keywords
© 2025 Xiaoming Zhang, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
With the rapid development of science and technology and the advancement of globalization, the professional construction of higher vocational colleges and universities is facing new challenges and opportunities. As a new mode of professional construction, professional group construction can effectively enhance the flexibility and adaptability of professional setting and improve the quality and efficiency of talent cultivation. Specialty group refers to a collection of several specialties with the same professional and technical basis or closely related specialties. This construction mode emphasizes the cross and integration between majors in order to form a new professional ecosystem. For higher vocational colleges and universities, the construction of specialty clusters is of great significance. It can improve the school’s response speed to market changes, improve the adaptability of talent training, but also can optimize educational resources and improve the overall school efficiency, which is of great significance in the higher vocational cultural and creative professions [1-2].
With the rapid development of social economy and people’s increasing attention to creative industries, the cultural and creative industry majors in colleges and universities are gradually emerging. The construction and practical research in this field has become an important part of college education. The specialty of cultural and creative industries covers a wide range of knowledge fields, including art, design, media and other disciplines. Based on the needs of their own industries, colleges and universities constantly update their professional courses, strengthen practical training and improve students’ professionalism. As the main place to cultivate talents, colleges and universities have important practical significance to strengthen the cultivation of professional talents in cultural and creative industries through the establishment of cultural and creative professional group teaching [3-6].
Literature [7] aims to discuss the curriculum development of cultural and creative majors, including curriculum issues, educational objectives, and curriculum construction. And expressed in the form of three diagrams and three tables, it emphasizes the overall context and relevance of curriculum reform to promote the quality of education in line with the international education accreditation norms. Literature [8] constructed an evaluation model using AHP method based on the professional group construction model of the three-class model. The results show that under the corresponding survey data, the professional group construction model connecting the industrial chain is the optimal group construction model, in addition, different professional group construction models have their rationality. Literature [9] explored the application of traditional Chinese thinking in design concepts, design methods and design education. Based on the case of cultural and creative design, it proves the effectiveness of the teaching model, which provides a new systematic approach to teaching and practicing in the field of cultural and creative design. Literature [10] presents a conceptual framework for analyzing diversity and opportunities in the cultural and creative industries as a result of specific decision-making and proposes three analytical foci. The results argue that the decision-making foci can lead to a shift in perspectives for understanding diversity and opportunities in the cultural and creative industries and improving them. Literature [11] explored the significance of building a new business professional community based on the integration of industry and education, and based on this, proposed the main measures for building a new business professional community based on the integration of industry and education. Literature [12] emphasized that professional education must break the original teaching framework based on a single specialty and reconstruct the professional group at the height of the whole industrial chain. Taking the specialty of architectural decoration engineering technology as the research object, it builds the talent training mode innovation of “combining curriculum certificate and competency standard” based on information technology for the integration of assembly building.
The research idea of this paper first systematizes the technological foundation of AIGC based on three perspectives: conceptual ontology, content production, and technological innovation. And centered on the endogenous value of AIGC, a four-energy education theory for AIGC is proposed. Combined with the ability of deep learning algorithms in perception, cognition, imitation and generation, four application modes of AIGC+ virtual teachers, teaching assistants, personalized learning and teaching content are constructed. The development status of AIGC technology applied to the field of art and design is studied, and the means and methods of application are analyzed. Through the distribution of questionnaires to determine the current more popular four AIGC tools “Midjourney, Stable Diffusion, Fabire, Vizcom”, based on the user’s ease of use experience to complete the four tools to assist in the art design of the effect of comparative experiments, found that Stable Diffusion is more suitable for the field of art design, and its generation method and characteristics are used to provide practical basis for subsequent intelligent education.
AIGC technology [13], as an advancement after PGC (Professionally-Generated Content), UGC (User-Generated Content), and AGC (AI-assisted Generated Content), automates through AI algorithms the generation and editing of data and multimedia content (e.g., images, audio, and video), marking the shift of the main body of content creation from people and institutions to AI, and realizing the technological leap from fuzzy retrieval to precise positioning. The foundation of Aigc technology is shown in Figure 1. In senior art and design professional education, AIGC provides an innovative solution to solve the complexity of 3D modeling and enhance the creativity challenges of architectural design, and deepens students’ understanding of the art and creative design process through AI-generated customized video analysis.
Basic overview of Aigc technology
GPT (Generative Pre-trained Transformer), which is a prominent representative of the AIGC field, is an autoregressive generative model based on the Transformer architecture proposed by OpenAI [14]. Its core principle covers pre-training, fine-tuning mechanism, multi-layer stacked Transformer structure, and positional encoding.
Before processing the raw text and image inputs, the model first transforms the text into vector form through Tokenization and Embedding, and at the same time performs feature extraction on the images to obtain their specific feature representations. This preprocessed information is then fed into the Transformer architecture, which utilizes the self-attention mechanism of the internal encoder to capture the complex relationships between elements in the input sequence. And the ordering information of the elements is preserved by positional encoding, which allows the model to understand the sequence and learn the global dependencies and local features of the sequence. Then, in the decoder part, the model further processes this information through self-attention and encoder-decoder attention mechanisms, focusing on the interactions between the input sequences and the target sequences to generate accurate outputs. This process not only enhances the model’s ability to understand complex natural language, but also enables it to learn different abstract representations at multiple levels, providing new perspectives and possibilities for future educational development.
In the pre-training phase of the model, the masked self-attention mechanism is introduced, which prompts the model to master deeper semantic connections between text and images by predicting missing information in text and images, laying a solid foundation for subsequent tasks. Subsequently, in the autoregressive generation phase, the model gradually constructs the output sequence relying on the previously generated text and image features, ensuring the coherence of the content and the appropriateness of the context, and fully considering the contextual information of the generated sequence. The subsequent fine-tuning session further refines the model to fit specific application scenarios, such as Q&A and text creation, so that it can effectively migrate the knowledge gained in the pre-training. The final output is transformed into a probability distribution through a Softmax layer to select the most likely vocabulary and image features to accomplish the task.
The deep learning architecture constructed by the entire model by fusing text and image information through a series of accurately planned steps is shown in Fig. 2, and it provides an effective way to process complex language comprehension and generation tasks. This efficient processing reflects the model’s ability to efficiently model and generate multimodal data, and also establishes the foundation for the diverse output of educational content.
GPT technical framework
Combining the current development of AIGC technology and the endogenous value of AIGC, this study proposes the AIGC four-energy education theory, that is, AIGC helps learners realize comprehensive development from low energy to high energy, from single energy to multiple energy, from multiple energy to super energy, and from super energy to heterogeneous energy through the participation in learning and personalized education, the facilitation of cross-disciplinary learning and comprehensive skill cultivation, the reinforcement of in-depth learning and the enhancement of higher-order thinking ability, and the expansion of cognitive boundaries and innovative thinking modes by means of auxiliary tools. The structure of AIGC’s four-energy theoretical model is shown in Figure 3. Starting from the definition of teaching and learning needs, AIGC assists learning and personalized education, facilitates interdisciplinary learning and comprehensive skill cultivation, strengthens in-depth learning and higher-order thinking ability, and expands cognitive boundaries and innovative thinking modes through the four-energy education, and combines the presentation of teaching and research technology to carry out The practice of teaching and researching is combined with the presentation of teaching and researching technology to finalize the innovation of education paradigm.
AIGC four-energy theoretical model architecture
Through personalized education, AIGC technology is used to help learners move from low proficiency levels to high proficiency levels quickly. By analyzing learners’ study habits, knowledge acquisition, and feedback, AIGC is able to improve the adaptability of teaching materials and customize learning content and teaching strategies for each student. For beginners, AIGC can recommend basic and detailed teaching materials, establish an interactive teaching process, and gradually improve students’ comprehension and application skills through practice and feedback on digital content.
Through the AIGC learning platform, a student who is new to art and design disciplines can receive personalized vocabulary, grammar, and listening and speaking training according to his or her own learning progress and interests, thus significantly improving the level of design and creation in a short period of time. In other words, AIGC personalizes the content and methods of teaching according to the needs and strengths of each learner.
At this stage, the AIGC Big Model serves as a tool that not only replaces traditional teaching tools, but also changes the teacher-centered approach to knowledge delivery. Through personalized and effective interaction and feedback mechanisms, the learning experience is active and deep in order to achieve the state transition from low energy to high energy and the cultivation of learning interest.
AIGC technology is used to help learners develop from a single skill to multiple mature skills; AIGC can integrate knowledge from various fields to provide learners with a wider range of learning resources to help them master multiple skills and develop comprehensive skills. For example, a student who is good at mathematical geometric analysis can, through the guidance of AIGC, start to learn how to apply mathematical knowledge such as geometric patterns and the golden section in mathematics to art design, and gradually develop interdisciplinary problem-solving skills.
At this stage, students use the big model to promote the expansion of the boundaries of collaborative learning, realize the integration of cross-disciplinary heterogeneous knowledge, expand the knowledge tree of skills, and broadly explore the “encyclopedia of mankind”, so as to achieve the effect of cross-cultural learning and cross-collaboration of multiple disciplines. Teachers can reasonably adopt GPT-enabled multimodal teaching resource construction methods in teaching practice, further promote the digitalization of multimodal teaching resources, and empower students with more timely, richer, and more exciting teaching content to enhance their learning and research capabilities.
Through deep learning and higher-order thinking skills development, AIGC technology is used to help learners develop from a state of multiple mature skills to a state of ultra-high competence. AIGC has the ability to deliver a high-quality, specialized body of knowledge that guides learners to delve deeper into a specific field and enhance their creativity, discernment, and problem-solving abilities. Through the application of large modeling technology, AIGC can also simulate real-world complex problems for students to solve, or provide experimental conditions through advanced simulation labs for students to conduct scientific research. For example, an animation design student was able to enhance their professional ability and innovative thinking in the field of animation design by creating freely through the virtual space lab.
At this level, AI is beginning to change the nature of teaching, focusing more on students’ deep and active participation in teaching practice, and on students’ ability to learn and think deeply and collaboratively across disciplines. At present, ChatGPT big language model has launched a multi-person document collaboration mode, Midjourney and other animation creation models can also be created through the new asset group (Assets) to serve multi-person multi-scene collaboration, teachers can use these tools to cultivate students’ ultra-high innovative thinking and technical skills.
AIGC has the ability to expand students’ cognitive boundaries and innovative modes of thinking, and then aid them in developing from a state of ultra-high ability to a state of extraordinary ability. The former is the acquisition, fulfillment, and enhancement of abilities based on the existing knowledge system of human beings, while the state of heteronormativity refers to the technological leap beyond the existing capabilities of human beings.The new contents created by AIGC, such as anti-physical aesthetics, intellectual illusion aesthetics, and infinite convergence aesthetics, are a great subversion of the existing presentation of the laws of the physical world, and the construction of imagination and aesthetics in the process of education and teaching. For example, students utilize AIGC tools to learn the principles of biology and apply them to sustainable design to create new design works.
In addition, the AIGC Educational Grand Model can be used in teaching to create new scenarios and tasks that were previously unimaginable, forming a learning scenario in which physical space and cyberspace are intertwined. Students can “visit” Cleopatra’s palace in a classroom the size of a square inch through the simulation space constructed by AIGC, or enter the inside of a whale’s skeleton to examine its structure. The use of technology goes far beyond the basic routines of displaying lecture notes on a projection screen and entering assessment scores into a digital database, expanding the boundaries of educators’ and students’ thinking. These teaching and learning experiences are continuously improved through real-time feedback and assessment. On this basis, AIGC can also assist learners in tapping into their latent cognitive abilities and developing unprecedented “psychic powers” to cope with the challenges and changes of the future society.
For AIGC to truly play a driving role in the field of media arts education, it needs to be deeply integrated with all kinds of segmented arts education scenarios. AIGC provides personalized learning forms, diversified content generation, and digitalized interaction forms for higher vocational art and design professional education. Especially, the basic ability of deep learning algorithms in the directions of perception, cognition, imitation, and generation supports the creative productivity of AIGC, forming the underlying technical support for the application scenarios of AIGC-enabled higher vocational art and design professional education. Scenario-oriented applications for art and design professional education mainly include four types of AIGC+ virtual teachers, AIGC+ teaching assistants, AIGC+ personalized learning, and AIGC+ teaching content as shown in Figure 4 [15].
Application scenario and production mechanism of AIGC in art design education
In the traditional “teacher-student” teaching model, the teacher is the authoritative transmitter of knowledge, but this model is faced with existing problems and drawbacks such as uneven teacher strength, the difficulty of individualized teaching, and the difficulty of growth of teachers’ heavy mechanical work. The intervention of AIGC technology demonstrates the possibility of transforming the traditional teaching model into a “teacher-student-computer” model. AIGC is able to increase the likelihood of achieving better teaching and learning by supplementing teacher resources, enabling large-scale personalized instruction, and reducing the burden on teachers, among other features.
This is reflected in the following two points [16]:
First, large-scale tailoring
AIGC makes large-scale tailored instruction possible. AIGC, natural language processing, and other technologies are used to develop virtual tutors that simulate real teachers and provide personalized guidance and feedback to students.
Secondly, realizing the “teacher-student-machine” teaching mode.
AIGC enables the teaching mode to shift from the traditional teacher-student teaching mode to the “teacher-student-machine” human-computer cooperative teaching mode. Combined with flipped classroom, XR and other technologies to stimulate students’ learning motivation, it provides students with a richer and more immersive learning experience, enhances the teaching effect through a variety of methods and strategies, and deepens students’ learning experience and understanding.
With the iterative progress of AIGC technology in terms of arithmetic, algorithms, model training, etc., AIGC based on real-time automated generation of content can realize real-time interaction with people, which can perceive the surrounding people, objects, and environment in real time and interact with them, and the most typical application representative is the humanoid robot. Empowered by AIGC, humanoid robots can become humanoid robot teaching assistants and virtual digital human companions for learners.
First, humanoid robot teaching assistant
Empowered by AIGC, humanoid robots can be transformed into “artists” to assist and facilitate learners to complete creative tasks.
Second, virtual digital human contextual companion
Through the contextual accompaniment of humanoid robots, AIGC can enhance the overall experience of learners in emotional, cognitive, interactive, and other educational processes, enhance learning motivation, and reduce learning pressure.
AIGC’s features, such as instant feedback and personalized learning, not only provide new possibilities for art education, but they can also be broken down into two types of personalized service scenarios for teachers and students.
First, AIGC personalized learning for the teacher side
AIGC generates primary movie scripts and film scripts through text, image descriptions and preset creative styles based on specific content training models in specific art and design education fields. Using technologies such as reinforcement learning and diffusion learning to generate video dubbing, automatic narration and sound cloning, it serves teachers in the content of art and design education courses such as theater and film literature, film and television editing, and animation production.
Secondly, AIGC Personalized Learning for Student Side
Based on AIGC, personalized teaching can be conducted to help learners customize personalized learning plans, inspire guided thinking, and conduct real-time Q&A. At the same time, AIGC can also provide real-time feedback to students when they are confused in the process of art creation, which can strengthen the learning effect and improve students’ subjective expression of art knowledge.
AIGC’s powerful automatic content generation capability makes art creation gradually intelligent, such as the use of generative adversarial network (GAN) and other technologies, which can realize the automated creation of text creation, image generation, video editing, and music synthesis in a variety of media. And this art education application scenario will be implemented in the art teaching content field, such as game development, art exhibition, art and culture IP, and art creation. So how to maximize the innovative teaching of AIGC in art education courses, so that students can better achieve the learning goals and adjust learning strategies to improve learning outcomes, is the key to enhance the effectiveness of art education teaching.
First, AIGC + Game Development
AIGC’s automatic content generation provides technical support for interactive scenarios such as game platforms and original game IPs, making it possible for learners to create more immersive and interactive game works.
Second, AIGC+Art Exhibition
The emergence of AIGC makes the art exhibition scene break the traditional “human-object-field” interaction connection, and provides learners with the opportunity to break through the time-space distance and combine the virtual scene with the real environment in the learning of the professional course of art exhibition and performance. It can also realize the effect of displaying contents in various modes of interaction and create immersive experience of art performance works.
Thirdly, AIGC+Art and Culture Creation
With the rise of the national tide wind, AIGC helps the digital creative presentation of the national tide culture through the automatic generation of content and other multimodal capabilities, realizes more attractive art and cultural works through the application scene of virtual-real fusion and multi-sensory interactive experience, and realizes the artistic conveyance of the cultural content, so as to enable the new generation to obtain spiritual resonance and cultural resonance.
This survey on the application of AIGC technology in higher vocational art and design education was conducted in the context of China’s efforts to promote the innovation of talent cultivation and education system of art and design professional groups and the increasingly obvious trend of the application of artificial intelligence technology in education. Therefore, the main purposes of this questionnaire survey include:
To understand the development status of AIGC technology application in higher vocational art and design education. To grasp the main ways of the application of AI technology in higher vocational art and design education. To summarize the problems existing in the application of AIGC technology in art and design education in higher vocational colleges.
Combining the viewpoints of review studies and theoretical studies, it has been summarized that the problems in the learning and creation of art and design professionals are concentrated and highly overlapping, and there is a preliminary grasp of the specific distribution of the problems in the application of AIGC technology in art and design education in higher vocational colleges and universities. After the questionnaire was formulated, first, 50 individuals were randomly selected within the target institutions for trial testing. The results of the pre-survey showed that the reliability test of the survey statistics was relatively satisfactory, with the value of the reliability index Clonbach
In terms of survey object selection, given that the research object of this paper is higher vocational art and design professionals, it is mainly divided into three categories: teachers, students, and art and design practitioners. Restricted by geographical conditions, this survey study selected three comprehensive colleges and universities in G city as the source of survey respondents. In terms of grade distribution, the respondents of this survey were selected to be teachers and students of art and design majors in all grades from freshman to junior. Based on this, the selection of survey respondents is characterized by multi-level, extensive, high credibility, and strong representativeness. A total of 500 questionnaires were distributed, including 300 questionnaires for art and design students and 286 questionnaires for teachers. Teachers: 100 copies, with 100 copies validly collected. There were 100 copies available for art and design practitioners, and only 82 were validly collected. The total effective recovery rate was 93.6%, of which female respondents slightly outnumbered male respondents by 54%.
In order to further deepen the understanding of the current application status of AIGC technology in art and design education in higher vocational colleges and universities, this section synthesizes quantitative analysis and qualitative analysis methods to summarize and summarize the application status.
Whether or not they have participated in AIGC-related art design work or projects The statistical results of the questionnaire on whether or not they have been involved in AIGC-related art and design work or projects are shown in Figure 5. The results show that practitioners, teachers, and students are arranged in descending order, indicating that the proportion of practitioners who have participated in AIGC-related art and design is the highest, followed by teachers, and the proportion of the student group is the lowest. However, in terms of the specific proportion of the student group, as high as 63.99% of the students have had AI art design experience or experience, and this proportion indicates that student participation in AIGC art design has been generalized on a certain scale. Knowledge of AIGC technology and its applications The survey respondents’ understanding of AIGC technology and its applications, the survey statistics are shown in Figure 6. ABCDE in the figure represents in order:
No understanding of the technology and how it is applied at all. Understand part of the AIGC technology but know nothing about its application. Partial understanding of the AIGC technology and partial understanding of its applications. In-depth understanding of AIGC technology, but no knowledge of its application. In-depth understanding of AIGC technology, its application has its own unique insights. The results showed that the student body was “partially aware of AIGC technology and partially aware of its applications.” The highest percentage (41.2%) indicates that most of the students are currently cognizant of AIGC in order to start the transition from the book stage, to the application stage. The consistency between the teacher group and the student group indicates that the teacher group has a certain mastery of the application of AIGC technology in art and design from theory to practice, but the degree is not too deep. Among the practitioners, the proportion of “in-depth understanding of AIGC technology, but ignorance of its application” is the highest (37.91%), but at the same time, the proportion of the option of “in-depth understanding of AIGC technology and having their own unique insights into its application” can also reach 30.21%, indicating that the practitioner group has made a certain breakthrough in the cognition and practice of AIGC technology applied to art design. Attitude towards the application of AIGC technology to assist art and design education in higher vocational colleges and universities For the survey respondents’ attitudes toward the application of AIGC technology-assisted art and design education in higher vocational colleges and universities, the statistical results are shown in Figure 7. ABCD in the figure represents in turn:
AIGC technology can’t be combined with art and design education yet. AIGC technology assists in accomplishing part of art and design education. AIGC technology will be fully applied to art and design education. The integration of AIGC technology with art and design education is kept under observation. The results show that “AIGC technology will be fully utilized in art and design education” has the highest percentage of respondents, with students contributing the most to the ratio. However, the proportional contribution values of students (55.07%), teachers (45%), and practitioners (53.74%) are all relatively close to each other, indicating that the majority of the respondents are optimistic and open to the application of AIGC technology to assist art and design education in higher education institutions. “AIGC technology-assisted completion of a portion of art and design education” was the next highest.
Have you ever participated in AIGC related art design work or projects
Survey respondents’ knowledge of AIGC technology and its applications
Attitude of applying AIGC technology to art design education in colleges and universities
Summarizing the results of the above questionnaire, the following conclusions were obtained:
Promote the popularization of AIGC application knowledge in art and design education. The popularization and application of a new technology in the real world can only be realized if a large enough number of users have the necessary “knowledge” of the new technology, so that the wide application of the new technology can have a realistic basis. The same applies to the application of AIGC technology in art and design education in higher vocational colleges and universities, and the cultivation of “attitude and interest” is the basis for the wide application of AIGC in the field of art and design in colleges and universities. Promote the practice of AIGC technology in the field of art and design education. There is a significant interaction between “cognition” and “practice and participation”, and they are mutually reinforcing. Similarly, the enhancement of the cognition and practice of the new technology, combined with the continuous maturity of the scale and popularization of the application of AIGC, will prompt more and more art designers and practitioners to move from cognition to practice, paving the way for the popularization of AIGC technology in the field of art and design teaching in colleges and universities. Developing good attitudes and interest in the application of AIGC technology. There is a significant interaction between “attitude and interest” and “practice and participation”, and they are mutually reinforcing. The level of users’ attitude and interest is closely related to whether they decide to apply the new technology in their own learning, teaching and work. In general, the application of AIGC technology in art and design education in vocational colleges is the interaction between the three dimensional variables of “cognition”, “practice and participation”, and “attitude and interest”, and continues to develop. The quantitative growth of each dimension can bring positive feedback to the other dimensions, and AIGC technology is driven by this positive feedback to accelerate the popularization and generalization.
Based on the questionnaire above, AIGC tools were selected from the market for practical application in the field of art and design. The experimental subjects are students majoring in art and design at a higher vocational college. At present, the commonly used AIGC tools on the market include Fabire, Vizcom, Midjourney, Stable Diffusion, etc. Different AIGC tools will produce different effect programs for the same content due to their unique algorithms and generation characteristics. The experiment focuses on exploring the generation effect of AIGC tools when different techniques are adopted for the same art design object. In addition, it also focuses on the effective cooperation between “tools” and “designers”, and evaluates the ease of use of the tools in the process of using, so as to make a comprehensive comparison and screening.
During the experiment, the experimental data are recorded for subsequent comparison and analysis. The data sources can be divided into three parts:
Record the generation time, the number of times generated, the number of unmet samples and the percentage of each subject who completed the design task by text generation, image generation and Lora-generated diagrams on different tools. At the end of the experiment, a user experience questionnaire was distributed to subjects majoring in art and design in higher vocational colleges to provide feedback on the subjective feelings of controllability, generation stability, and ease of learning and practice of these four AIGC tools in the process of applying them to the use of art design. The generated samples of each AIGC tool were scored for user satisfaction to assess the usability of AIGC technology applied in the field of art design.
During the above experimental process, systematic data were collected from each subject who utilized four AIGC tools, Fabire, Vizcom, Midjourney and Stable Diffusion, to complete the design tasks in the form of text generation, image generation, and plug-in graph generation, respectively.
The data related to average task length (A/min), average number of generation times (B), number of unmet samples (C), and percentage of unmet samples (D/%) are shown in Table 1. Specifically, longer task completion times usually imply that the tool is more difficult to use or that there is a mismatch between the current generation method and the desired art and design content, resulting in lower generation quality. In addition, the number of generation times reflects to some extent the subjects’ satisfaction with the generated content. The higher the number of unmet samples and the lower the number of generation times tend to indicate that the quality of the generated art and design content is less satisfactory.
Statistics of average task completion time and generation times
| Type \ Tool | Fabire | Vizcom | Midjourney | Stable Diffusion | ||||
|---|---|---|---|---|---|---|---|---|
| Text | Graphics | Text | Graphics | Text | Graphics | Text | Graphics | |
| A/min | 10.82 | 8.31 | 8.24 | 15.21 | 10.04 | 13.24 | 14.21 | 12.84 |
| B | 5.07 | 7.67 | 8.71 | 12.26 | 5.57 | 7.81 | 7.58 | 10.16 |
| C | 6.00 | 3.00 | 8.00 | 2.00 | 11.00 | 7.00 | 10.00 | 1.00 |
| D/% | 25 | 16 | 32 | 6 | 49 | 22 | 55 | 9 |
An in-depth analysis of the data in the table reveals that the text-generated approach is generally higher than the image-generated approach in terms of the length of time to complete the task and the number of unmet samples. This result clearly indicates that the application of the text generation approach in the field of art and design is somewhat limited by its relatively low efficiency and accuracy, which makes it less suitable for design tasks in this field. In addition, Midjourney’s data feedback in the experiment performed relatively poorly in comparison with the other three tools, where the number of unattained samples amounted to 11 and 7 times for text and images, respectively, which was the highest among the four tools. Comparatively, Vizcom and Stable Diffusion had the lowest number of unmet samples in the image generation method, showing high generation quality and applicability. This suggests that they have a supporting role in the field of art and design and can meet the needs of designers more effectively.
The optimal generation samples (subjective optimal) of each participant in the experiment were collected, and since the text generation approach is not applicable to the art and design domain, only 98 optimal picture generation samples were evaluated in this paper. Using the Likert 5-order scale method to allow the respondents to evaluate the satisfaction scores of the effect solutions generated by each AIGC tool, the data feedback obtained is shown in Table 2. The samples of generated art design solutions were mixed and disorganized, and the three points of innovativeness, accuracy and clarity were used as the assessment criteria for scoring and evaluation, with the score ranging from 0 to 5, where 0 means very dissatisfied and 5 means very satisfied.
Generate sample user satisfaction ratings
| Type \ Tool | Fabire | Vizcom | Midjourney | Stable Diffusion |
|---|---|---|---|---|
| Accuracy | 2.68 | 3.51 | 3.86 | |
| Innovativeness | 3.75 | 1.84 | 3.69 | |
| Articulation | 2.94 | 3.46 | 3.42 | |
| Comprehensive satisfaction | 3.12 | 3.09 | 3.61 |
An in-depth analysis of the data in the table reveals that the solutions generated by the Midjourney and Fabire tools perform the best in terms of innovativeness, and despite the presence of inaccurate identification of some parts, deviations in material and color matching, and other details, the solutions generated by these tools can be used to inspire designers with creative assistance, which is a good application. In terms of the accuracy of the generated solutions, Vizcom tool’s performance is most in line with the original design input, although there is a large gap in terms of innovation and other tools, but it can be applied to the process of conversion from sketches to renderings, the traditional design process, due to the visibility of the sketches is not intuitive enough, can not be used as a program directly to the A party to confirm. Through Vizcom tools, you can reduce the time consumed by program modification, thus improving design efficiency.
The user experience data were mainly derived from a subjective evaluation questionnaire for subjects of 98 art and design students in higher vocational colleges and universities, focusing on the four Fabire (AIGC1), Vizcom (AIGC2), Midjourney (AIGC3), and Stable Diffusion (AIGC4) tools under the art and design domains to generate solutions that are controllability, generation process stability and learning difficulty. The results of the subjects’ questionnaire evaluations are shown in Figure 8, where the subjects gave the least dissatisfied feedback to the Stable Diffusion tool, indicating that it was superior to the other three tools in terms of controllability and stability. As for the ease of learning and practicing, Midjourney performed better, showing that it is easier for users to get started, compared to Stable Diffusion, which has the highest learning difficulty.
Questionnaire evaluation of subjects
Summarizing the experimental data and questionnaire feedback, it can be concluded that although Stable Diffusion is slightly higher in terms of hardware requirements and learning difficulty, its powerful controllability, trainability, and excellent generative effects give it a significant advantage in the field of art and design. The tool has been able to fulfill most of the needs of designers, providing rich inspiration and enhancing design efficiency. However, due to the complexity and diversity of the data, AIGC is not yet able to fully understand the design purpose in small domain content learning, so the generated art design solutions are more for reference, and the control of the final design results still depends on the professionalism of the designers. In addition, the other three tools also show their own unique advantages, namely Fabire can effectively promote thinking in the early stage of design, Vizcom can simplify the process of sketching conversion, and Midjourney shows excellent performance in the field of graphic design.
After an in-depth comparison and analysis of the four AIGC tools Fabire, Vizcom, Midjourney and Stable Diffusion. Finally, the Stable Diffusion tool was selected as the preferred tool in the field of art and design professional education in higher education institutions. The content generated by this tool shows a higher degree of fit in terms of art design needs, so in practical application, Stable Diffusion will be utilized to assist the preliminary work of art design, including providing creative inspiration and speeding up the conversion from sketches to renderings, with a view to obtaining more innovative and satisfying art design results.
The emergence of AIGC has brought a brand new education mode such as autonomous teaching and personalized learning to the education of senior art and design majors, and has promoted the exploration of the creation of art and design education. Through the study, it is found that the most problem that users have in the application of AIGC technology and art design education is that the rational thinking of the machine does not fit well with the emotional thinking of people. At this stage, AIGC technology is more in the way of educational aids, participating in all aspects of teaching and design, bringing a new interactive form of transformation for art and design talents, a new new lecture logic enhancement, a new method of work creation and a new way of display. These new changes affect the development of art and design education and teaching, and subvert the traditional teaching mode, the interaction of educational products, creative thinking and other aspects of the cognition. Looking into the future, the arrival of strong artificial intelligence is inevitable, and the change of the role of art design workers in the creative process also has a strong research value. How to tap into the unique spirit of art design creators, how to better utilize AIGC in art design creation, and other issues are of great significance to the cultivation of art design talents.