Projects & Datasets
My research builds large-scale, real-world datasets and benchmarks for driving automation — lane-keeping assist, human takeovers, control transitions, driver behavior, and safety alerting. Below is the full set of projects. Explore more of my work on Google Scholar, Hugging Face, ResearchGate, and GitHub.
OpenLKA
The first large-scale open dataset of Lane Keeping Assist from 62 production vehicle models under real-world driving, pairing decoded CAN-bus logs with synchronized 1080p video.
ADAS-TO
15,000+ real-world ADAS takeover events from 327 drivers across 22 brands — multimodal and vision-language-annotated — capturing the moment drivers reclaim control.
BATON
A multimodal benchmark for bidirectional human–automation control transitions: 136.6 h of naturalistic driving with video, CAN, radar and GPS, and three prediction tasks.
DriveDNA
A benchmark for driving-style identification — 4,121 drives from 465 drivers, 975 h — that isolates driver-specific behavior from vehicle, route and environment confounds.
DriveMotion
A multi-source benchmark for driver body-motion forecasting: 400 h of in-cabin skeleton motion (133 keypoints/frame, privacy-reduced) with an 8 s → 4 s prediction task.
VLAlert
Observe Before You Alert: a vision–language framework for adaptive driver alerting (SILENT / OBSERVE / ALERT), with VLAlert-Bench — 192,892 one-second labeled ticks from six datasets.