KITScenes Multimodal: European Urban Driving Data with 360° Sensor Coverage

A high-fidelity European urban autonomous-driving dataset built for the FiftyOne platform. Each frame contains synchronized data from a full robotaxi sensor suite, including nine global-shutter cameras, seven long-range lidars, and three 4D imaging radars. The dataset is packaged by Voxel51 and was last updated on June 8, 2026.

Use Cases

• Train multimodal sensor fusion models based on synchronized camera, lidar, and radar data.
• Develop and test HD-map-based localization algorithms based on the provided Lanelet2 map labels.
• Benchmark depth estimation models based on the projected lidar depth information.
• Research future trajectory prediction for autonomous vehicles based on the provided future ego path data.
• Evaluate instance segmentation and object detection models based on the image instance predictions.

Strengths

• High-fidelity sensor suite includes nine cameras for 360° coverage, seven lidars, and three 4D radars per frame.
• Includes production-grade Lanelet2 HD-map labels, projected lidar depth, and future ego path data.
• Synchronized multimodal capture from a full robotaxi sensor suite.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.
• Data may reflect geographic bias inherent to its collection in European urban environments.

Provenance

Source

Voxel51 (packager), original source is the KIT-MRT (KITScenes Multimodal) dataset.

Collection Method

Synchronized capture from a robotaxi sensor suite in urban environments.

Freshness

Last updated 2026-06-08 11:51:46; freshness should be verified.

Geography

European urban areas.