Snapshot: Towards Application-centered Models for Pedestrian Trajectory Prediction in Urban Traffic Environments

TL;DR

This paper introduces Snapshot, a real-time, robust pedestrian trajectory prediction model tailored for urban traffic, outperforming existing methods and validated within an autonomous driving system.

Contribution

We propose Snapshot, a novel modular neural network for pedestrian prediction that improves accuracy, efficiency, and robustness, and establish a new benchmark based on Argoverse 2.

Findings

Snapshot reduces ADE by 8.8% compared to state-of-the-art.

It demonstrates scalability and real-time performance.

The model is robust to varying motion histories.

Abstract

This paper explores pedestrian trajectory prediction in urban traffic while focusing on both model accuracy and real-world applicability. While promising approaches exist, they often revolve around pedestrian datasets excluding traffic-related information, or resemble architectures that are either not real-time capable or robust. To address these limitations, we first introduce a dedicated benchmark based on Argoverse 2, specifically targeting pedestrians in traffic environments. Following this, we present Snapshot, a modular, feed-forward neural network that outperforms the current state of the art, reducing the Average Displacement Error (ADE) by 8.8% while utilizing significantly less information. Despite its agent-centric encoding scheme, Snapshot demonstrates scalability, real-time performance, and robustness to varying motion histories. Moreover, by integrating Snapshot into a modular autonomous driving software stack, we showcase its real-world applicability.