Interactive Color Style Recommendation And Fast Rendering Method For Indoor Scenes

In recent years,intelligent interior design has gradually become the focus of people’s attention and research.Interior design often needs professional designers to complete,which requires not only rich design knowledge,but also professional design software.It takes a lot of time to achieve the desired design effect.For the furniture combination determined in the indoor scene,it can help designers and users to complete the design better to recommend a reasonable set of furniture color matching and realize the quick preview of the renderings.In this paper,an interactive color theme recommendation and fast rendering method for indoor scene is studied.The purpose is to use the designer’s prior knowledge to assist users to match the indoor scene furniture colors interactively,and quickly synthesize the design drawings to achieve the purpose of design preview.The main work and innovation of this paper are as follows:(1)This paper proposes a furniture color recommendation algorithm based on a priori knowledge,and summarizes the material-furniture-color association mapping by studying the association relationship between furniture colors and materials in interior scenes,so as to build a database of common materials for the furniture studied in this paper.The experimental results show that the material-furniture-color association mapping and the furniture color recommendation algorithm based on the a priori knowledge are good at assisting users to complete the interior furniture color matching and greatly improve the efficiency of interior design.(2)A multi-layer gray image joint interactive coloring algorithm based on deep learning is proposed.By selectively coloring and fusing the pre rendered gray scene image database,the fast rendering preview of interior design drawings is realized.The algorithm is based on deep learning coloring network.By pre rendering 8-10 gray level images of indoor scene,combined with virtual user point algorithm and multi-layer gray level image coloring algorithm,it can quickly draw and preview design drawings.Under the same equipment conditions,compared with the rendering time of professional design software,which is often several minutes or even hours,the algorithm proposed in this paper is hundreds of times more efficient than professional software in interactive rendering while ensuring the rendering quality of the design,which can significantly improve the efficiency of interior scene design.(3)This paper proposes an interactive scene color style fast generation system.The system designs a clear interactive interface and simple interactive operation,integrates the furniture color recommendation algorithm based on the prior knowledge and the multi-layer gray image joint coloring algorithm based on deep learning,and realizes the interactive coloring of furniture in the indoor scene and the fast synthesis preview of the design drawings of the indoor scene.The system proposed in this paper has good usability and effectiveness.Compared with professional design rendering software,the average interaction times of the system is 1-4 times on the premise of ensuring the quality of indoor scene synthesis.The operation is simple and efficient,and the ordinary users can complete the interior design independently.The design results not only follow the user’s subjective interaction concept,but also combine the professional designer’s furniture appearance Color matching prior,the system is real and reliable,with good user interaction.