PulseAugur
EN
LIVE 22:49:06

New VZCrash dataset details 31,000 vehicle crashes

Researchers have introduced VZCrash, a new dataset containing over 31,000 real-world vehicle crashes recorded using Inertial Measurement Unit (IMU) telemetry. This dataset, collected from a fleet of over 73,000 vehicles in the United States, includes detailed acceleration, angular velocity, and GPS data. The researchers used this dataset to benchmark various crash detection models, highlighting the critical role of data scale in training effective models for real-world deployment. AI

IMPACT Provides a large-scale dataset for training and evaluating AI models in vehicle crash detection.

RANK_REASON The cluster contains a research paper detailing a new dataset.

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 English(EN) · Tommaso Bianconcini, Henrique Pi\~neiro Monteagudo, Aurel Pjetri, Tomaso Trinci, Leonardo Taccari ·

    VZCrash: A Large-Scale IMU Dataset of Ego-Vehicle Crashes

    arXiv:2606.06074v1 Announce Type: new Abstract: We introduce VZCrash, the largest publicly available dataset of real-world vehicle collision data featuring Inertial Measurement Unit (IMU) telemetry. The dataset contains more than 31,000 validated crashes and 158,000 negative samp…

  2. arXiv cs.CV TIER_1 English(EN) · Leonardo Taccari ·

    VZCrash: A Large-Scale IMU Dataset of Ego-Vehicle Crashes

    We introduce VZCrash, the largest publicly available dataset of real-world vehicle collision data featuring Inertial Measurement Unit (IMU) telemetry. The dataset contains more than 31,000 validated crashes and 158,000 negative samples, including hard cases and distractors. Each …