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

2025-06-27

AI Summary

To address the insufficient robustness of pedestrian intention and trajectory prediction in unstructured traffic scenarios, this paper introduces IDD-PeD—the first large-scale pedestrian dataset tailored to India’s complex driving environments. IDD-PeD captures realistic challenges including illumination variations, severe occlusions, and unsignalized intersections, and provides dual-level annotations: high-level intentions (e.g., “about to cross”, “pausing to observe”) and low-level actions (e.g., “initiating motion”, “decelerating”). Built upon high-fidelity video capture and multi-round manual verification, it establishes a unified evaluation benchmark. Quantitative evaluation reveals substantial performance degradation of state-of-the-art models: intention recognition accuracy drops by up to 15%, and trajectory prediction MSE increases by as much as 1208. IDD-PeD fills a critical gap in fine-grained behavioral modeling for unstructured settings and serves as a foundational resource and new benchmark for enhancing prediction reliability of autonomous driving systems in complex real-world environments.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Large pretrained models with web data

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

Problem

• Predict pedestrian behavior in unstructured traffic conditions
• Address dataset gaps for robust pedestrian intention and trajectory models
• Evaluate performance drop in existing methods on complex environments

Innovation

• Introduces Indian driving pedestrian dataset for unstructured traffic
• Provides comprehensive annotations for pedestrian behavior analysis
• Evaluates state-of-the-art prediction methods showing performance gaps