|
With
the rapid development of artificial intelligence
technology, painting style migration, as a cross field
between computer vision and art creation, provides new
possibilities for art education innovation. This study
explores the application of drawing style migration
technology in the teaching of computer and art
integration, aiming to solve the problems of low
efficiency of style learning, high threshold of
creation, and insufficient interdisciplinary integration
in traditional art teaching. By constructing a deep
learning-based drawing style migration teaching
framework and combining quantitative and qualitative
research methods, the effectiveness of the model in
enhancing students' artistic expression, depth of
technical understanding and creative thinking ability is
verified.
The
study adopts a controlled experimental design, selecting
students from an art college and dividing them into an
experimental group (integrated teaching) and a control
group (traditional teaching) for a 16-week teaching
experiment. The experimental group learns about
stylistic feature extraction and parameter tuning
through VGG19 and ResNet50 pre-training models, and
combines art history analysis with digital creation
practice. The research constructed a three-dimensional
assessment system of ¡°artistic performance-technical
mastery-learning effectiveness¡±, and quantitatively
analyzed the index of stylistic diversity, creativity
novelty, algorithmic comprehension accuracy and other
indexes of the students. The experimental results show
that the experimental group is significantly better than
the control group in terms of stylistic diversity,
technical mastery and creative efficiency. The
qualitative analysis further shows that the style
migration technology promotes students' in-depth
understanding and creative expression of art styles
through the triple effects of ¡°algorithmic lens¡±,
¡°creative gas pedal¡± and ¡°thought converter¡±. The
innovation of this study lies in the following: the
proposed algorithmic lens, the creative accelerator, and
the thought transformer.
The
innovations of this study include: proposing a
progressive teaching model of
¡°deconstruction-restructuring-creation¡±, which reduces
the technical learning curve; developing a lightweight
style migration tool for educational applications, which
supports real-time parameter adjustment and effect
feedback; and revealing the intrinsic mechanism of the
integration of art and computational thinking. Practice
shows that the drawing style migration technology not
only enhances teaching efficiency, but also expands the
boundaries of artistic creation, providing a
generalizable paradigm for interdisciplinary art
education. Future research can further explore the
adaptive learning path and the design of creative
environment that integrates reality and fiction.
¡¡ |
