Den-TP: A Density-Balanced Data Curation and Evaluation Framework for Trajectory Prediction

TL;DR

This paper introduces Den-TP, a data-centric framework for curating and evaluating trajectory prediction datasets, addressing scenario density imbalance to improve model robustness, especially in safety-critical high-density situations.

Contribution

Den-TP presents a density-aware data curation and evaluation framework that balances scenario densities, reducing dataset size while enhancing robustness in high-density scenarios.

Findings

Dataset size reduced by 50% without performance loss

Improved robustness in high-density scenarios

Revealed long-tail failure modes with density-conditioned evaluation

Abstract

Trajectory prediction in autonomous driving has traditionally been studied from a model-centric perspective. However, existing datasets exhibit a strong long-tail distribution in scenario density, where common low-density cases dominate and safety-critical high-density cases are severely underrepresented. This imbalance limits model robustness and hides failure modes when standard evaluations average errors across all scenarios. We revisit trajectory prediction from a data-centric perspective and present Den-TP, a framework for density-aware dataset curation and evaluation. Den-TP first partitions data into density-conditioned regions using agent count as a dataset-agnostic proxy for interaction complexity. It then applies a gradient-based submodular selection objective to choose representative samples within each region while explicitly rebalancing across densities. The resulting subset reduces the dataset size by 50\% yet preserves overall performance and significantly improves robustness in high-density scenarios. We further introduce density-conditioned evaluation protocols that reveal long-tail failure modes overlooked by conventional metrics. Experiments on Argoverse 1 and 2 with state-of-the-art models show that robust trajectory prediction depends not only on data scale, but also on balancing scenario density.