Pedestrian Intention and Trajectory Prediction in Unstructured Traffic Using IDD-PeD

TL;DR

This paper introduces a new Indian pedestrian dataset for unstructured traffic environments, highlighting the challenges in predicting pedestrian behavior and evaluating current models' performance drops in such complex scenarios.

Contribution

The paper presents a novel dataset capturing unstructured traffic conditions and provides comprehensive annotations, enabling improved pedestrian intention and trajectory prediction research.

Findings

State-of-the-art methods perform significantly worse on the new dataset.

Trajectory prediction models show up to 1208 MSE increase in unstructured environments.

Evaluation reveals substantial performance gaps in existing models for complex traffic scenarios.

Abstract

With the rapid advancements in autonomous driving, accurately predicting pedestrian behavior has become essential for ensuring safety in complex and unpredictable traffic conditions. The growing interest in this challenge highlights the need for comprehensive datasets that capture unstructured environments, enabling the development of more robust prediction models to enhance pedestrian safety and vehicle navigation. In this paper, we introduce an Indian driving pedestrian dataset designed to address the complexities of modeling pedestrian behavior in unstructured environments, such as illumination changes, occlusion of pedestrians, unsignalized scene types and vehicle-pedestrian interactions. The dataset provides high-level and detailed low-level comprehensive annotations focused on pedestrians requiring the ego-vehicle's attention. Evaluation of the state-of-the-art intention prediction methods on our dataset shows a significant performance drop of up to $\mathbf{15\%}$, while trajectory prediction methods underperform with an increase of up to $\mathbf{1208}$ MSE, defeating standard pedestrian datasets. Additionally, we present exhaustive quantitative and qualitative analysis of intention and trajectory baselines. We believe that our dataset will open new challenges for the pedestrian behavior research community to build robust models. Project Page: https://cvit.iiit.ac.in/research/projects/cvit-projects/iddped