Accepted inputs
- Point-cloud sequences
- Synchronized camera and LiDAR data
- Pre-annotations requiring refinement
- Client-hosted 3D annotation projects
3D PERCEPTION DATA
LiDAR and 3D Cuboid Annotation
3D annotation requires consistent spatial extent, orientation and identity—not just boxes in a camera view. We work to client-defined coordinate, class and visibility rules and review cuboids across synchronized frames where supported by the selected platform.
SCOPE
What this service covers
3D cuboid placementPosition, dimensions and orientationObject classes and attributesSequence-level track consistencyCamera and point-cloud cross-checkingExisting cuboid correction
Annotation approach
- Ground-plane and object-extent alignment
- Heading and orientation checks
- Class and attribute assignment
- Sparse-point handling
- Temporal consistency review
Common edge cases
- Sparse or distant objects
- Partial scans and occlusion
- Overlapping point clusters
- Ambiguous object extent
- Stationary versus moving-object identity
QUALITY ASSURANCE
Review that follows the guideline
Quality checks are selected for the data type and acceptance criteria. Metrics and review coverage depend on the project, platform and agreed workflow.
Cuboid fit and orientation reviewClass and attribute checksTrack continuity validationCross-view consistency checksExport and coordinate-schema validation
Applications
- Autonomous mobility
- Robotics
- Mapping
- Smart infrastructure
- Industrial perception
Output formats
- Platform-native 3D exports
- KITTI-compatible structures where specified
- Client-defined JSON
- Synchronized frame manifests
What to include in your brief
- Sensor configuration and coordinate conventions
- Class and attribute taxonomy
- Cuboid extent and orientation rules
- Tracking requirements
- Platform, format and acceptance criteria
START WITH A CALIBRATION BATCH
Reduce uncertainty before production.
Share your guidelines, sample characteristics, volume, preferred platform and timeline. We’ll assess the scope and propose a representative calibration step.
Discuss your dataset