Constructing High-Quality Livable Cities: A Comprehensive Evaluation of Urban Street Livability Using an Approach Based on Human Needs Theory, Street View Images, and Deep Learning

Driven by the United Nations’ Sustainable Development Goal (SDG 11), the construction of high-quality livable cities has emerged as a central issue on the global agenda. However, existing research primarily focuses on optimizing physical functions, neglecting the dynamic hierarchical nature and emotional experiences of residents’ needs. This study, employing Guangzhou’s Tianhe District as an empirical case, proposes an innovative framework that integrates Maslow’s Hierarchy of Needs theory, the Method of Empathy-Based Stories (MEBS), and deep learning technology for the first time. It constructs a dynamic assessment model of “needs-streetscape elements-spatial quality”, systematically analyzing the livability characteristics and driving mechanisms of high-density urban streets. Tianhe District’s street spaces exhibit the common issue of “functional-experiential imbalance” faced by high-density cities. Furthermore, different streetscape elements in the city demonstrate significant variability in satisfying different hierarchical demand dimensions, with strong sequential relationships among these hierarchies. Adjusting and optimizing the relationships between elements can result in the creation of higher-quality street spaces that meet higher-level needs. The research findings provide differentiated renewal pathways for tropical high-density cities, offer methodological support for global urban governance under the SDG 11 objectives, and indicate directions for improving street quality in urban regeneration practices.