WTS: A Pedestrian-Centric Traffic Video Dataset for Fine-grained Spatial-Temporal Understanding

TL;DR

This paper introduces the WTS dataset, a comprehensive traffic video dataset focusing on pedestrian and vehicle behaviors for fine-grained spatial-temporal understanding, supporting advanced traffic safety and autonomous driving research.

Contribution

The paper presents the WTS dataset with multi-perspective annotations, 3D gaze data, and an LLM-based evaluation metric, filling a gap in pedestrian-centric traffic video analysis.

Findings

WTS dataset includes over 1.2k annotated traffic events.

Introduces LLMScorer for evaluating video captioning accuracy.

Establishes a benchmark for dense video-to-text tasks in traffic scenarios.

Abstract

In this paper, we address the challenge of fine-grained video event understanding in traffic scenarios, vital for autonomous driving and safety. Traditional datasets focus on driver or vehicle behavior, often neglecting pedestrian perspectives. To fill this gap, we introduce the WTS dataset, highlighting detailed behaviors of both vehicles and pedestrians across over 1.2k video events in hundreds of traffic scenarios. WTS integrates diverse perspectives from vehicle ego and fixed overhead cameras in a vehicle-infrastructure cooperative environment, enriched with comprehensive textual descriptions and unique 3D Gaze data for a synchronized 2D/3D view, focusing on pedestrian analysis. We also pro-vide annotations for 5k publicly sourced pedestrian-related traffic videos. Additionally, we introduce LLMScorer, an LLM-based evaluation metric to align inference captions with ground truth. Using WTS, we establish a benchmark for dense video-to-text tasks, exploring state-of-the-art Vision-Language Models with an instance-aware VideoLLM method as a baseline. WTS aims to advance fine-grained video event understanding, enhancing traffic safety and autonomous driving development.