ACCIDENT: A Benchmark Dataset for Vehicle Accident Detection from Traffic Surveillance Videos

TL;DR

ACCIDENT is a comprehensive benchmark dataset with real and synthetic traffic accident videos, designed to evaluate various accident detection tasks in CCTV footage under different data scenarios.

Contribution

The paper introduces ACCIDENT, a new dataset and benchmark for vehicle accident detection, including multiple tasks and diverse baselines, addressing challenges in CCTV footage analysis.

Findings

ACCIDENT dataset contains over 4,000 annotated clips.

The benchmark evaluates temporal, spatial, and collision type classification.

Baseline methods show the dataset's difficulty and room for improvement.

Abstract

We introduce ACCIDENT, a benchmark dataset for traffic accident detection in CCTV footage, designed to evaluate models in supervised (IID and OOD) and zero-shot settings, reflecting both data-rich and data-scarce scenarios. The benchmark consists of a curated set of 2,027 real and 2,211 synthetic clips annotated with the accident time, spatial location, and high-level collision type. We define three core tasks: (i) temporal localization of the accident, (ii) its spatial localization, and (iii) collision type classification. Each task is evaluated using custom metrics that account for the uncertainty and ambiguity inherent in CCTV footage. In addition to the benchmark, we provide a diverse set of baselines, including heuristic, motion-aware, and vision-language approaches, and show that ACCIDENT is challenging. You can access the ACCIDENT at: https://accidentbench.github.io