基于多模态时空数据融合的城市空间通用表征学习方法、装置、终端及存储介质
By fusing multimodal spatiotemporal data to generate a general urban representation, the problem of poor applicability of multimodal data is solved, and a unified representation and applicability for various urban analysis tasks are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN UNIV
- Filing Date
- 2026-04-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies have not yet been able to effectively transform multimodal spatiotemporal data into a universal urban representation applicable to various urban analysis tasks, resulting in data distribution discrepancies.
By acquiring multimodal spatiotemporal data of the target city, a single-view representation of each spatial unit is generated. Multi-view fusion is performed using an attention mechanism and a preset algorithm to generate a multi-view fusion representation matrix, which is then globally aggregated to obtain a general representation of the city.
It achieves a unified representation of multimodal spatiotemporal data, improves the applicability of general urban representations, and can be applied to diverse urban analysis tasks, such as population distribution prediction, travel flow prediction, and environmental quality prediction.
Smart Images
Figure CN121980249B_ABST