Objects Waterfront urban spaces with unique geographical advantages and distinctive morphological features are essential spatial carriers for showcasing the distinctive characteristics, superior quality, and pleasant living environment of the city. This research focuses on how to activate and reuse waterfront spaces, and how to better utilize and guide the creation of waterfront space interfaces that align with the visual attention patterns of the public. The research identifies two limitations in previous studies: Unclear impact factors and specific effects of waterfront space interfaces on the diverse dynamic behaviors of people; lack of a constructed mechanism linking dynamic attention patterns of the crowd to the morphological characteristics of waterfront space interfaces. Therefore, the research aims to explore the following two questions: 1) clarify the key impact factors and specific influences of morphological changes in waterfront space interfaces on the diverse dynamic attention of the crowd; 2) construct a linkage mechanism between indicators regarding the morphology of waterfront space interface and those regarding dynamic visual attention of the crowd, so as to provide a basis for subsequent research and practical guidance.

Methods Taking the Nanjing Hexi waterfront area as an example, this research employs techniques such as oblique photography measurement, real-world data image collection, urban interface morphology analysis, and physiological eye movement indicators monitoring to establish the correlation mechanism between “spatial interface” and “visual attention”, thus elucidating the preferences of viewers under different movement modes and their correlation with various indicators regarding the morphology of waterfront space interface. The specific research steps are as follows. 1) Collection of waterfront space samples. The research adopts the oblique photography measurement technology to obtain high-precision 3D point cloud models and create a spatial sample library, and selects the waterfront road along the main axis of the research area as the primary experimental path for image collection. 2) Experimental design and operation. The research divides spatial video samples into standard units of 4-minute duration for different movement rates, creating a total of 11 samples for subsequent experiments and data extraction. 3) Quantification of dynamic attention data. The research employs physiological eye-tracking devices to collect data on changes in the visual attention of the crowd in different movement modes towards the interface in waterfront space samples, involving eye movement indicators regarding dynamic attention such as fixation rate, return rate, and average diameter of left and right pupils. 4) Quantification of data on waterfront space interface. The research combines data on physiological eye movement with the landing points in the test area, and uses the ArcGIS platform to quantitatively calculate various spatial interface indicators, involving a number of indicators regarding spatial morphology of waterfront interface such as interface occupancy, interface hierarchy quantity, contour undulation frequency, and contour undulation amplitude. 5) Analysis of “attention − interface” data. The research conducts correlation analysis and establishes a regression model for the two-dimensional indicators of “dynamic attention of the crowd” and “waterfront space interface”, and clearly defines the linkage mechanism between the spatial morphological characteristics of waterfront interface and dynamic attention of the crowd, in an effort to explore the variation patterns of waterfront interfaces and the impact of waterfront interface indicators on various dynamic attentions.

Results The research reveals that the visual attention of waterfront space visitors is positively influenced by interface occupancy, interface hierarchy quantity, contour undulation frequency, and contour undulation amplitude. Specifically, the fixation rate is significantly affected only by interface occupancy in a positive manner. The return rate is influenced positively by both contour undulation frequency and contour undulation amplitude. As for the average diameter of left and right pupils, the regression results indicate significant impacts from both interface hierarchy quantity and contour undulation frequency. Overall, the creation of waterfront space should focus on the well-designed vertical interface hierarchy of buildings, a high proportion of visual occupancy by buildings, and outstanding waterfront interfaces with significant variations in building heights and frequencies. Furthermore, the influence of “spatial interface” on “visual attention” indicators varies significantly under different movement modes, requiring specific analysis for each movement mode.

Conclusion Based on the findings of this research, for waterfront parks in already densely built areas, interface occupancy, interface hierarchy quantity, contour undulation frequency, and contour undulation amplitude can be regarded as crucial indicators and metrics for assessing the quality of visual attention experience in waterfront spaces. Establishing the correlation between population attention and interface indicators enables differentiated design. Particularly, given the current context of existing urban development, in the subsequent evaluation and control processes, differentiating the assessment of visual attention quality along paths of different movement modes can contribute to the creation of more refined waterfront tour spaces and experience. The research results successfully construct the correlation mechanism between urban waterfront interface morphology and dynamic population attention, providing robust support for the orderly guidance of waterfront urban interfaces, as well as related spatial layout and planning decisions.