Region-Point Joint Representation for Effective Trajectory Similarity Learning

TL;DR

RePo is a novel trajectory similarity learning method that jointly encodes region-wise and point-wise features, capturing spatial context and movement patterns to significantly improve accuracy over existing methods.

Contribution

The paper introduces RePo, a new approach combining region-wise and point-wise features with cross-attention for enhanced trajectory similarity modeling.

Findings

RePo achieves 22.2% higher accuracy than SOTA methods.

Joint encoding of spatial and movement features improves similarity assessment.

Contrastive training with hard negatives enhances model robustness.

Abstract

Recent learning-based methods have reduced the computational complexity of traditional trajectory similarity computation, but state-of-the-art (SOTA) methods still fail to leverage the comprehensive spectrum of trajectory information for similarity modeling. To tackle this problem, we propose \textbf{RePo}, a novel method that jointly encodes \textbf{Re}gion-wise and \textbf{Po}int-wise features to capture both spatial context and fine-grained moving patterns. For region-wise representation, the GPS trajectories are first mapped to grid sequences, and spatial context are captured by structural features and semantic context enriched by visual features. For point-wise representation, three lightweight expert networks extract local, correlation, and continuous movement patterns from dense GPS sequences. Then, a router network adaptively fuses the learned point-wise features, which are subsequently combined with region-wise features using cross-attention to produce the final trajectory embedding. To train RePo, we adopt a contrastive loss with hard negative samples to provide similarity ranking supervision. Experiment results show that RePo achieves an average accuracy improvement of 22.2\% over SOTA baselines across all evaluation metrics.