Diving Deeper Into Pedestrian Behavior Understanding: Intention Estimation, Action Prediction, and Event Risk Assessment

TL;DR

This paper introduces a comprehensive benchmark for pedestrian behavior understanding, focusing on intention estimation, action prediction, and event risk assessment, analyzing models and data modalities across these tasks.

Contribution

It defines new evaluation metrics and benchmarks for pedestrian behavior tasks, and provides a comparative analysis of state-of-the-art models on these tasks.

Findings

Different data modalities impact model performance.

Intention estimation and action prediction are complementary.

Model agreement varies across tasks and scenarios.

Abstract

In this paper, we delve into the pedestrian behavior understanding problem from the perspective of three different tasks: intention estimation, action prediction, and event risk assessment. We first define the tasks and discuss how these tasks are represented and annotated in two widely used pedestrian datasets, JAAD and PIE. We then propose a new benchmark based on these definitions, available annotations, and three new classes of metrics, each designed to assess different aspects of the model performance. We apply the new evaluation approach to examine four SOTA prediction models on each task and compare their performance w.r.t. metrics and input modalities. In particular, we analyze the differences between intention estimation and action prediction tasks by considering various scenarios and contextual factors. Lastly, we examine model agreement across these two tasks to show their complementary role. The proposed benchmark reveals new facts about the role of different data modalities, the tasks, and relevant data properties. We conclude by elaborating on our findings and proposing future research directions.