.The product modeling design is the concrete manifestation of realizing the unified identification goal of the corporate image. It can shape and disseminate the corporate image, create individual brands, and win in the fierce market competition. The product modeling design needs to serve the overall image design of the enterprise, with product design as the core, around people’s needs for products, and to the greatest extent possible to meet the needs of individuals and society to obtain a general sense of recognition. It changes people’s way of life and improves the quality and level of life. Different pieces of information need to be combined when designing a product. To achieve the best effect, we aim to study the application of image fusion algorithm based on digital media technology in modeling design. Digital media technology is a technology that combines abstract numbers, physical media, and computer technology. It is a generation of digital media technology for digital audio, digital video, digital film, computer animation, virtual reality, and so on. A compressed sensing multifocus image fusion algorithm based on non-subsampled shearlet transform is proposed. It improves the efficiency of the fusion algorithm without affecting the visual effect. We analyze the shape design of wearable smart products, summarizes the characteristics of product shape design, as well as the aesthetic value is shown, and summarizes the entry point of product shape design. The experimental results of this paper show that the proportion of choosing the classic style is 55.3%, which shows that people are more comfortable with the classic style, so they are more likely to buy. Health monitoring is the mainstream function of wearable products, and its proportion is as high as 71.3%, indicating that people are more concerned about health issues.