Collaborative Innovation of Poster Design and CAD Based on Gradient Descent Algorithm

Abstract

As big data technology continues to advance and find broader applications, its impact on the design sector is steadily growing. This article aims to delve into how poster design and computer-aided design (CAD) systems, fueled by big data, can foster collaborative innovation via gradient descent algorithms. Initially, the significance of big data in poster design becomes evident, as it offers designers precise market direction and a deeper understanding of user preferences through rigorous analyses of user behaviour data and market trends. Additionally, the article sheds light on the particular use of the gradient descent algorithm in image matching, highlighting its ability to assist designers in swiftly identifying materials that closely align with their design concepts from extensive image databases. In addition, the article also verified the significant advantages of image-matching methods based on stochastic parallel gradient descent (SPGD) in poster design and CAD collaborative innovation through empirical research. It explored how to optimize algorithm performance in practical operations further to improve design efficiency and quality. This innovative method not only injects new vitality into poster design but also provides strong support for upgrading and improving CAD systems.