MGTraj: Multi-Granularity Goal-Guided Human Trajectory Prediction with Recursive Refinement Network

2025-09-11

AI Summary

Existing target-guided trajectory prediction methods predominantly adopt a two-stage paradigm—coarse-grained target prediction followed by fine-grained trajectory completion—overlooking the modeling potential of intermediate temporal granularities, thereby limiting multi-granularity synergy. To address this, we propose MG-Refine, a Multi-Granularity target-guided Recursive Refinement network, the first to systematically integrate multi-granularity modeling into the target-guided framework. MG-Refine enables progressive trajectory refinement from coarse to fine scales via cross-granularity feature sharing, recursive iterative optimization, and weight-sharing mechanisms. Additionally, it incorporates a velocity prediction auxiliary task and a Transformer-based recursive architecture to enhance spatiotemporal consistency and representational capacity. Extensive experiments on ETH/UCY and Stanford Drone datasets demonstrate that MG-Refine significantly outperforms state-of-the-art baselines, achieving new SOTA performance among target-guided approaches.

Technology Category

• Computer Vision: Motion & Tracking
• Multiagent Systems: Multiagent Planning
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate human trajectory prediction is crucial for robotics navigation and autonomous driving. Recent research has demonstrated that incorporating goal guidance significantly enhances prediction accuracy by reducing uncertainty and leveraging prior knowledge. Most goal-guided approaches decouple the prediction task into two stages: goal prediction and subsequent trajectory completion based on the predicted goal, which operate at extreme granularities: coarse-grained goal prediction forecasts the overall intention, while fine-grained trajectory completion needs to generate the positions for all future timesteps. The potential utility of intermediate temporal granularity remains largely unexplored, which motivates multi-granularity trajectory modeling. While prior work has shown that multi-granularity representations capture diverse scales of human dynamics and motion patterns, effectively integrating this concept into goal-guided frameworks remains challenging. In this paper, we propose MGTraj, a novel Multi-Granularity goal-guided model for human Trajectory prediction. MGTraj recursively encodes trajectory proposals from coarse to fine granularity levels. At each level, a transformer-based recursive refinement network (RRN) captures features and predicts progressive refinements. Features across different granularities are integrated using a weight-sharing strategy, and velocity prediction is employed as an auxiliary task to further enhance performance. Comprehensive experimental results in EHT/UCY and Stanford Drone Dataset indicate that MGTraj outperforms baseline methods and achieves state-of-the-art performance among goal-guided methods.

Problem

• Predicting human trajectories with multi-granularity goal guidance
• Integrating intermediate temporal granularity into trajectory modeling
• Recursively refining trajectory proposals from coarse to fine levels

Innovation

• Multi-granularity recursive refinement network
• Transformer-based feature integration
• Velocity prediction auxiliary task