Color Matching Generation Algorithm for Animation Characters Based on Convolutional Neural Network

Area of Science:

• Computer Science
• Artificial Intelligence
• Digital Media Arts

Background:

• The Chinese animation industry is a rapidly growing sector with increasing global recognition.
• Traditional color matching techniques for animation characters lack innovation and efficiency.
• Automated color generation is crucial for advancing animation production.

Purpose of the Study:

• To innovate and improve the color matching generation for animation characters using advanced algorithms.
• To develop a novel method that enhances the accuracy and efficiency of character colorization.
• To address limitations in traditional color matching within the animation industry.

Main Methods:

• Utilized convolutional neural networks (CNNs) for innovative color matching.
• Employed Generative Adversarial Network (GAN), Deep Convolutional Generative Adversarial Network (DCGAN), and VGG models.
• Developed and applied a novel LMV-ACGAN research method with a multiscale discriminator theory.

Main Results:

• The proposed LMV-ACGAN model demonstrated a lower collapse rate compared to other models.
• Achieved superior color matching results for animation characters.
• Observed improvements in color matching correlating with increased CNN utilization.

Conclusions:

• The LMV-ACGAN method offers a significant advancement in automated color matching for animation characters.
• This approach enhances the quality and efficiency of color generation in the animation industry.
• The study provides a foundation for future research in AI-driven animation.