Estimating Automobile Crash Characteristics from Images using Deep Learning
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Crash characteristics such as crash velocity (Delta-V) and location of collision (LOC) are important determinants of the severity of the injury sustained by an occupant of an accident vehicle. Based on the predicted severity levels of injury, insurance companies can estimate the claim’s cost and better plan their financial reserves. We present a promising approach for accurately predicting Delta-V and LOC using deep learning methods, without the need for a forensic crash reconstruction. We constrain the study to small passenger vehicles and to front and rear collisions with crash velocities under 96 kph. We first develop and refine our image processing and deep CNN architectures using images created by using vehicle crash simulation software. Using a k-fold cross-validation approach, our methods are able to predict the crash velocity of simulated collisions (108 images) with a MAE of 3.41 kph (MAPE of 8.2%). Similarly, a multiple task learning CNN is able to predict Delta-V of real-world collisions (310 images) with a MAE of 4.19 kph (MAPE of 16.2%) and classify the LOC with 92% accuracy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it