Multimodal Driver Response Data Across Diverse Road Scenarios

EmoRoad provides anonymized clip and raw data capturing psychological, physiological, and behavioral human-subject responses in varied driving conditions. The 3.3 GB dataset was created by RCFCM Hong Kong and released as open access in April 2026. It integrates multiple sensor modalities to study driver states.

Use Cases

• Correlate physiological signals like heart rate with behavioral video clips to detect driver stress levels.
• Train multimodal fusion models to predict psychological self-report labels from concurrent time-series sensor data.
• Analyze behavioral response patterns across different driving scenario clips for human-machine interaction research.
• Benchmark emotion recognition algorithms using synchronized physiological and video recordings from real-world tasks.

Strengths

• Dataset size of 3.3 GB indicates substantial multimodal content.
• Data is fully anonymized with no personally identifiable information, easing ethical use.
• Released under the permissive CC-BY-4.0 license for open research.

Limitations

• Unknown participant count and row count limits reproducibility and statistical power assessment.
• Lack of published column details or sample data obscures the specific features and structure available.

Provenance

Source

RCFCM Hong Kong, via figshare.

Collection Method

Human-subject data collection in driving scenarios using multimodal sensors.

Time Range

null

Freshness

Last updated April 2026, indicating recent availability.

Geography

Data collection likely centered in Hong Kong, but not explicitly stated.