TUMTraffic-VideoQA: A Benchmark for Unified Spatio-Temporal Video Understanding in Traffic Scenes

TL;DR

TUMTraffic-VideoQA introduces a comprehensive dataset and benchmark for evaluating spatio-temporal reasoning in traffic scene videos, integrating multiple tasks to advance intelligent transportation research.

Contribution

It provides a new large-scale dataset with unified tasks and a baseline model, addressing the complexity of real-world traffic scene understanding.

Findings

The dataset contains 1,000 videos with diverse conditions.

Existing models show limitations on this challenging benchmark.

The TUMTraffic-Qwen baseline offers insights into fine-grained reasoning challenges.

Abstract

We present TUMTraffic-VideoQA, a novel dataset and benchmark designed for spatio-temporal video understanding in complex roadside traffic scenarios. The dataset comprises 1,000 videos, featuring 85,000 multiple-choice QA pairs, 2,300 object captioning, and 5,700 object grounding annotations, encompassing diverse real-world conditions such as adverse weather and traffic anomalies. By incorporating tuple-based spatio-temporal object expressions, TUMTraffic-VideoQA unifies three essential tasks-multiple-choice video question answering, referred object captioning, and spatio-temporal object grounding-within a cohesive evaluation framework. We further introduce the TUMTraffic-Qwen baseline model, enhanced with visual token sampling strategies, providing valuable insights into the challenges of fine-grained spatio-temporal reasoning. Extensive experiments demonstrate the dataset's complexity, highlight the limitations of existing models, and position TUMTraffic-VideoQA as a robust foundation for advancing research in intelligent transportation systems. The dataset and benchmark are publicly available to facilitate further exploration.