Robusto-2: 20 Dashcam Videos from Lima and NYC for Model Comparison

Robusto-2 contains 20 dashcam videos used in a research paper. The videos are split evenly between Lima, Peru and New York City, USA, with 10 videos from each region. The dataset was created by Artificio and is associated with a paper from 2026.

Use Cases

• Compare model performance across different geographic regions based on the split of 10 videos from Lima and 10 from NYC.
• Analyze urban traffic patterns and driver behavior using dashcam video footage.
• Benchmark object detection or action recognition models on real-world driving scenarios.
• Study domain adaptation or generalization challenges in computer vision using data from two distinct cities.

Strengths

• Provides a controlled comparison set with 10 videos from each of two distinct geographic regions (Lima and NYC).
• Derived from a larger set of 200 dashcam videos, suggesting a curated selection for specific research purposes.

Limitations

• Dataset size is small, containing only 20 videos, which may limit statistical power for some analyses.
• Column-level documentation and detailed metadata are absent, requiring manual inspection after download.
• Row count and file formats are unknown, complicating suitability assessment and preprocessing planning.

Provenance

Source

Artificio via Hugging Face.

Collection Method

Selected from a larger collection of 200 dashcam videos recorded in Lima, Peru and New York City, USA.

Freshness

Last updated 2026-06-23 15:09:59.

Geography

Lima, Peru and New York City, USA.