AI algorithms in visual communication design: enhancing design creativity and efficiency

At this stage, product packaging design is seriously homogenized, lacking unique visual elements and innovative styles, which affects consumers' memory and choices. This paper introduces AIGC (Artificial Intelligence Generated Content) visual communication technology, using automated generation and real-time feedback capabilities to generate design documents and improve brand recognition. It combines neural style migration and uses VGG (Visual Geometry Group)-19 deep neural network technology to extract tea cultural elements and modern design style characteristics and intelligently integrates them into tea packaging design to make the packaging meet the diversified market needs. The experimental results show that the cosine similarity range of the generated green tea packaging is 0.34–0.43. Compared with the packaging products of major e-commerce platforms, they have a lower similarity, which effectively reduces the packaging similarity problem. Compared with algorithms such as Transformer-based Image Synthesis, the experimental design performs the best in generation time, reaching 50 minutes, greatly reducing the packaging design time, and the calculation energy consumption is 240 W/h, better than other algorithms' consumption. AIGC combined with neural style transfer can enhance the creativity of packaging design and maintain competitiveness in the market environment, providing an effective solution for future packaging design.