Cross-View Urban Traffic Dataset: Synchronized Bicycle and Drone Videos from Germany

The Cross-View Urban Traffic Dataset (CVUTD) is a benchmark for cross-view urban traffic perception. It is built from synchronized ego-centric bicycle videos and aerial drone videos recorded at real urban intersections in Regensburg, Germany. The dataset was authored by prakharbh and last updated on Hugging Face in May 2026.

Use Cases

• Train models for cross-view identity matching between street-view and drone-view object tracks as described.
• Develop ego-to-BEV (bird's-eye-view) prediction models using aerial supervision as a benchmark task.
• Research multi-modal perception systems for autonomous vehicles or traffic monitoring based on synchronized video streams.

Strengths

• Dataset is explicitly designed as a benchmark for two linked computer vision tasks.
• Data is recorded from real urban intersections, likely providing realistic scenarios.
• Videos are synchronized, providing temporally aligned ground-truth across perspectives.

Limitations

• Description metadata is limited; actual data quality, size, and file formats require manual inspection after download.
• Row count and scale are unknown, which may limit suitability assessment for large-scale training.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

huggingface

Collection Method

Recorded from synchronized ego-centric bicycle videos and aerial drone videos.

Freshness

Last updated 2026-05-07 10:00:04; freshness should be verified.

Geography

Regensburg, Germany