Estimating crash details from multi-view vehicle photos
From Wrecks to Wisdom: Recovering Crash Mechanics from Real-World Multi-View Photos
Computer Vision and Pattern Recognition
Summary
Knowing how a car crash happened helps with safety, insurance, and injury prevention, but crash reports often miss important details. This paper shows how photos taken from different angles after a crash can be used to estimate key crash information like where the impact hit and how fast the car changed speed. The researchers trained a computer model on thousands of crash cases with many photos to predict these details. While some measurements are easier to predict than others, their method improves prediction accuracy and offers a way to use photos when official data is incomplete.
What this means in practice
- •For insurance adjusters: Use post-crash vehicle photos to estimate crash impact details for faster claim triage when standard data is missing.
- •For vehicle safety analysts: Estimate key crash parameters from accident photos to support injury modeling and improve safety evaluations in incomplete data cases.
Authors
Ondřej Valach, Václav Diviš, Ivan Gruber
Abstract
Estimating accident mechanics from real-world crashes is important for vehicle-safety analysis, injury modeling, crash-severity prediction, and operational workflows such as insurance claim triage. In standard crash records, key metadata such as impact configuration, principal direction of force, and change in velocity ($ΔV$) may be missing, delayed, or corrupted, while post-crash photographs are widely available and contain rich visual evidence of deformation. We study how much crash-mechanics information can be recovered directly from vehicle photos when structured signals are absent. We formulate crash understanding as supervised prediction from per-case multi-view photo sets. Targets include six Collision Deformation Classification (CDC) descriptors and the longitudinal and lateral components of reconstructed $ΔV$. Each photo is encoded by a shared visual backbone, and the resulting view-level features are fused into a case-level representation from which target-specific heads predict crash descriptors. Using 15.2k training cases from the Crash Investigation Sampling System, drawn from about 1.5M photos before filtering, together with 1.15k validation and 1.15k test cases, we define an evaluation protocol for vision-based crash descriptor estimation from incomplete multi-view evidence. Post-crash imagery alone provides usable signal for several non-trivial crash-mechanics descriptors, while weakly observable and long-tailed targets remain challenging. Within the compared training regimes, the selected joint-training recipe reduces mean absolute angular error for principal direction of force from 20.1 to 14.05 degrees and longitudinal $ΔV$ MAE from 8.04 to 7.45 km/h. Our work provides a reference point for future multimodal fusion with structured crash metadata.