Naturalistic driving data extraction and processing for studying driver head scanning behavior at intersections
Shrinivas Pundlik, Seonggyu Choe, Patrick Baker, Chen-Yuan Lee, Naser Al-Madi, Alex R. Bowers, Gang Luo · 2026
Read Naturalistic driving data extraction and processing for studying driver head scanning behavior at intersections by Shrinivas Pundlik, Seonggyu Choe, Patrick Baker, Chen-Yuan Lee, Naser Al-Madi, Alex R. Bowers, Gang Luo (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13547230.txt.
Description
Shrinivas Pundlik and co-authors built in 2026 a processing pipeline that mines ordinary driving recordings for how drivers scan with their heads at intersections, where about half of all crashes happen. Cars carried cabin and forward cameras plus speed and location logs. Software clipped the footage around intersections, read signs and maneuvers from the road view, and estimated head turns from the cabin view with a custom model built for wide angles.
Against 365 hand-checked intersection clips from six vehicles in three states, the pipeline spotted signage in 99 percent of cases and maneuvers in 95 percent. Estimated entry points landed within about a meter of the truth, and head pose error averaged under seven degrees. The authors aim at automated study of very large driving datasets.
Catalog record
| CREATOR | Shrinivas Pundlik, Seonggyu Choe, Patrick Baker, Chen-Yuan Lee, Naser Al-Madi, Alex R. Bowers, Gang Luo |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64443 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |