With the rapid development of the Internet and the Internet of Things, smart city and smart home are gradually penetrating into people's lives. Through the cognition of existing smart furniture models in the market, research is conducted on the problem that the wisdom of furniture is only reflected in the electrical appliances, while the wisdom of resting sofas is underdeveloped. The wisdom of the smart sofa model lies in its ability to understand the user's habits and preferences through deep learning and to realize the user's living needs in terms of movement or temporal characteristics. Using a combination of qualitative and quantitative analysis, the metaverse virtual model of intelligent furniture to realize the contradiction between user needs and real furniture defects. The results of the SPSS modeling experiments demonstrate the reliability and rationality of the blood pressure monitoring, heart rate monitoring, stand up assistance and childcare functions of the smart sofa. The innovative construction of a virtual model of a smart sofa provides thinking and data support for the development of smart furniture products.