Hu, Liyang (Southeast University), Gao, Kun (Chalmers University of Technology), Shao, Yichang (Nanjing University of Posts and Telecommunications), Shi, Xiaomeng (Southeast University), Ye, Zhirui (Southeast University)

Multi-View Tailored Tensor Completion for Spatiotemporal Traffic Data Imputation

Scheduled for presentation during the Regular Session "S39a-Data-Driven Optimization in Intelligent Transportation Systems" (FR-LM-T39), Friday, November 21, 2025, 10:50−11:10, Coolangata 3

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Abstract

Advanced sensing technologies have enabled multi-view observations of traffic dynamics, yet data missingness remains a significant challenge to reliable traffic monitoring. Unlike existing single-view imputation approaches, we propose Multi-View Tailored Tensor Completion (MVT2C), a framework integrating intra-view and inter-view modules. The intra-view module employs third-order tensor schemes to characterize high-dimensional low-rank properties within each view, while the inter-view module captures view-to-view relationships through subspace representation and accommodates view-specific discrepancies via structured column-sparse reconstruction error matrices. We formulate the integrated framework as a multi-block non-convex optimization problem, solved efficiently using an inexact augmented Lagrangian multiplier method. Experiments on two real-world datasets demonstrate the superiority of MVT2C across various missing scenarios compared to state-of-the-art baselines. Further analyses reveal that leveraging cross-view information not only improves robustness and stability but also yields affinity matrices that reflect both shared and unique view-specific features.