Three Steps to Multimodal Trajectory Prediction: Modality Clustering, Classification and Synthesis

TL;DR

This paper introduces a three-step multimodal trajectory prediction framework involving modality clustering, classification, and synthesis, which outperforms existing methods on standard benchmarks without using social or map data.

Contribution

The paper proposes a novel three-step framework for multimodal trajectory prediction that addresses limitations of previous methods and achieves state-of-the-art results.

Findings

Achieves 19.2% improvement in ADE on ETH/UCY dataset

Achieves 20.8% improvement in FDE on ETH/UCY dataset

Outperforms existing methods without social or map information

Abstract

Multimodal prediction results are essential for trajectory prediction task as there is no single correct answer for the future. Previous frameworks can be divided into three categories: regression, generation and classification frameworks. However, these frameworks have weaknesses in different aspects so that they cannot model the multimodal prediction task comprehensively. In this paper, we present a novel insight along with a brand-new prediction framework by formulating multimodal prediction into three steps: modality clustering, classification and synthesis, and address the shortcomings of earlier frameworks. Exhaustive experiments on popular benchmarks have demonstrated that our proposed method surpasses state-of-the-art works even without introducing social and map information. Specifically, we achieve 19.2% and 20.8% improvement on ADE and FDE respectively on ETH/UCY dataset. Our code will be made publicly availabe.