SceneSelect: Selective Learning for Trajectory Scene Classification and Expert Scheduling

TL;DR

SceneSelect introduces a scene-centric selective learning approach for trajectory prediction, dynamically routing inputs to expert models based on scene characteristics, significantly improving accuracy and generalization across diverse environments.

Contribution

The paper proposes SceneSelect, a novel scene-centric paradigm with unsupervised scene clustering and expert routing, enhancing trajectory prediction accuracy and adaptability without extensive retraining.

Findings

Outperforms single-model and ensemble baselines by 10.5% on average.

Effectively generalizes across datasets without joint retraining.

Demonstrates robustness on ETH-UCY, SDD, and NBA benchmarks.

Abstract

Accurate trajectory prediction is fundamentally challenging due to high scene heterogeneity - the severe variance in motion velocity, spatial density, and interaction patterns across different real-world environments. However, most existing approaches typically train a single unified model, expecting a fixed-capacity architecture to generalize universally across all possible scenarios. This conventional model-centric paradigm is fundamentally flawed when confronting such extreme heterogeneity, inevitably leading to a severe generalization gap, degraded accuracy, and massive computational waste. To overcome this bottleneck, rather than refining restricted model-centric architectures, we propose selective learning, a novel scene-centric paradigm. It explicitly analyzes the characteristics of the underlying scene to dynamically route inputs to the most appropriate expert models. As a concrete implementation of this paradigm, we introduce SceneSelect. Specifically, SceneSelect utilizes unsupervised clustering on interpretable geometric and kinematic features to discover a latent scene taxonomy. A highly decoupled classification module is then trained to assign real-time inputs to these scene categories, and a highly extensible, plug-and-play scheduling policy automatically dispatches the trajectory sequence to the optimal expert predictor. Crucially, this decoupled design ensures excellent generalization capabilities, allowing seamless integration with different off-the-shelf models and robust adaptation across new datasets without requiring computationally expensive joint retraining. Extensive experiments on three public benchmarks (ETH-UCY, SDD, and NBA) demonstrate that our method consistently outperforms strong single-model and ensemble baselines, achieving an average improvement of 10.5%, showcasing the effectiveness of scene-aware selective learning.