Autonomous Vehicles

Perception data with ASIL-tagged provenance.

LiDAR, camera, radar, and sensor-fusion datasets for autonomous driving perception, prediction, and planning. KITTI 3D + NuScenes-compatible formats. Every dataset ships with ASIL classification and an Ed25519-signed LQS cert your functional-safety team can verify.

Sensors
4
Camera · LiDAR · radar · fused-stream
Formats
3
KITTI 3D · COCO JSON · Parquet (tabular logs)
ASIL tags
A–D
ISO 26262 classification · cert-embedded
Anonymization
Auto
Face + plate blur pre-publication · attested
Functional-safety + privacy fields mapped into every cert
ISO 26262
ASIL A–D
SOTIF
ISO 21448
EU AI Act
Art. 10 aligned
GDPR
face + plate blur
Ed25519
signed cert
Featured datasets

Driving data with verifiable provenance.

Live marketplace listings filtered to autonomous driving. Every card shows signed LQS score, ASIL classification, scenario tags (weather, time of day, road type), and originality signal.

Tasks covered

Every perception + prediction task in the AV stack.

LiDAR point clouds
3D point-cloud datasets with 3D bounding-box annotations for vehicle, pedestrian, and cyclist detection at range. Calibration + intensity preserved.
formats · .pcd · .bin · KITTI
Camera perception
Multi-camera surround-view datasets with 2D/3D object detection, lane segmentation, and depth-estimation labels. Synchronized timestamps.
modes · mono · stereo · surround
Lane & road detection
Lane-marking segmentation, road-surface detection, and HD-map annotation datasets for path-planning and localization models.
schema · pixel + spline
Traffic sign & signal
Traffic-sign classification and traffic-light state detection across multiple countries and lighting/weather conditions.
coverage · EU · US · APAC
Sensor fusion
Synchronized LiDAR + camera + radar sequences with calibration matrices and temporal correspondences for end-to-end fusion models.
calib · intrinsic + extrinsic
Pedestrian behavior
Pedestrian trajectory, intent prediction, and interaction datasets for social-force and transformer-based prediction models.
horizon · 1–8s rollout
KITTI 3D · 3D bbox + calib COCO · JSON (camera only) Parquet · tabular sensor logs .pcd / .bin · point cloud NuScenes-compat · multi-sensor
What LabelSets adds

Cert fields designed for functional-safety review.

AV training data is regulated under ISO 26262, ISO 21448, and the EU AI Act. LabelSets certs carry the ASIL classification, scenario tagging, and anonymization attestation your safety case needs.

ASIL-tagged scenarios
Every AV dataset carries ISO 26262 ASIL A–D classification at the scenario level. Edge-case tagging (weather, night, construction) for SOTIF coverage.
field · asil_class · scenario_tag
Face + plate anonymization
Automated face and license-plate detection + blur pre-publication. Sellers attest to completeness; post-facto flags trigger cert revocation.
field · anon_attested · timestamp
Ed25519-signed provenance
Every cert carries a public-key signature + fingerprint. Buyers verify at /verify any time. Drops into ISO 26262 Part 6 software-safety evidence packages.
fingerprint · aa4c070af907e2ea
FAQ

Questions safety teams actually ask.

KITTI 3D (with 3D bounding boxes and camera-LiDAR calibration files) is the primary supported format. COCO JSON is supported for camera-only detection, and Parquet is available for large-scale tabular sensor logs. NuScenes-compatible multi-sensor bundles are also accepted.
Camera (monocular and stereo), LiDAR point clouds (.pcd, .bin), radar, and multi-modal sensor fusion datasets. Many datasets include synchronized sensor streams with timestamp alignment and full calibration matrices.
Yes, if you hold the rights and the data has been properly anonymized (faces and license plates blurred). Upload, pass verification and the automated anonymization check, set your price, and earn 85% of every sale.
Yes. Sellers tag datasets by weather condition (rain, fog, snow, night) and scenario type (intersection, highway, construction zone, unprotected left). Scenario tags are embedded in the signed cert to support ISO 21448 (SOTIF) safety-case coverage mapping.
The LQS cert drops into ISO 26262 Part 6 software-safety evidence packages — the Ed25519-signed audit chain, asil_class, and scenario tagging give your functional-safety team the dataset-quality artifact they need. Buyers should confirm suitability with their own safety case.

Browse all autonomous datasets.

Live marketplace filtered by LQS score, ASIL class, sensor mix, and scenario tags. Or list your own fleet data and earn 85% per sale.

Browse datasets → Sell your dataset
Related categories
Computer Vision NLP / Text Audio & Speech Medical Imaging Financial Data