Creative Design and Visual Effect Simulation of IoT Dynamic Interfaces by Fractal Algorithms and Artificial Intelligence

Contemporary dynamic interface designs for consumer IoT devices are constrained by several pressing issues, including excessive homogenization, insufficient real-time adaptability, and low generative efficiency. These constraints limit the ability to meet growing user expectations for personalized interactions and high-fidelity virtual experiences. To address these issues, this study proposes Puff-Net, an intelligent generation framework that overcomes such limitations through the synergistic integration of fractal algorithms and neural style transfer. First, multi-scale artistic textures are created by exploiting fractal self-similarity, while an end-to-end aesthetic control is achieved using a dynamic optimization module that incorporates gradient constraints and fractal dimension regularization. Second, a dual-branch feature architecture is designed to separately extract device state information and user preference style features, which are adaptively recombined to enable the dynamic visual expression of a holographic avatar. Finally, the framework ensures privacy and security while meeting the stringent requirements for multi-resolution display and real-time interaction in Internet of Things scenarios. Experimental results indicate that the proposed framework effectively enables dynamic interface generation across applications such as smart home systems and wearable devices. Through the coordinated use of fractal generation and style transfer, Puff-Net surpasses the limitations of traditional template-based design approaches, offering a lightweight solution that harmonizes artistic expression with functional responsiveness. Overall, this study presents a novel theoretical and technical foundation for integrating holographic representations within the IoT ecosystem, paving the way for the next generation of intelligent and personalized user interfaces in consumer electronics.