Interior Design Layout Optimization Based On Convolutional Neural Network

In interior design, due to the complex spatial structure and the interweaving of multifunctional requirements, how to effectively optimize the layout of furniture and decorative elements and improve space utilization and visual comfort has become a core challenge for designers. To this end, this paper proposes an optimization research method based on Convolutional Neural Network (CNN). Firstly, convolutional neural networks are used to extract features of interior layout sample data in order to analyze the layout patterns between different furniture and spatial elements. Secondly, an interior design layout generation model is constructed to learn the spatial planning methods of excellent designs through convolutional neural networks, and user needs and spatial functions are considered in the generation process. Finally, based on multi-layer convolution and transfer learning strategies, the model is able to efficiently handle various types of indoor environments and provide reasonable layout optimization solutions. The experimental results show that the proposed CNN model achieves a maximum space utilization rate of 78%, which is significantly better than algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO); in terms of visual satisfaction indicators, the CNN model has a layout symmetry score of 0.9 and a color uniformity score of 0.2, both of which are superior to other algorithms. The overall results show that the CNN model has significant advantages in space utilization, visual aesthetics, and functional adaptability, effectively solving optimization problems in diversified layout design.