Delivery format
Point cloud format for robot training data
Point cloud is useful for 3D scene geometry, object reconstruction, LiDAR/depth capture, navigation perception, and manipulation planning. Define coordinate frame, units, sensor intrinsics and extrinsics, timestamps, segmentation or object labels where available, and source RGB/depth references before reviewing samples so you can verify that delivery matches the training pipeline.
Quick facts
- Common containers
- PCD (PCL library, 2010), PLY (Stanford, 1994), LAS/LAZ (LiDAR community, 2003), USDZ (Pixar/Apple 2018, used by Aria / Project Aria).
- Sensor sources
- Velodyne / Ouster / Hesai LiDAR; Intel RealSense / Microsoft Azure Kinect / Stereolabs Zed RGBD — 6+ major sensor families in active use.
- Frame conventions
- ROS REP-103 (right-handed, +X forward) is most common in robotics; OpenGL / Unity / Unreal use 3 different conventions — define explicitly.
- Required fields
- Coordinate frame, units, sensor intrinsics + extrinsics, timestamps, segmentation labels, source RGB/depth references.
Comparison
| Format choice | Strength | Risk |
|---|---|---|
| Point cloud | 3D scene geometry, object reconstruction, LiDAR/depth capture, navigation perception, and manipulation planning | Needs exact schema agreement before capture |
| Raw files | Fast supplier export | High buyer cleanup burden |
| Custom schema | Matches internal pipeline | Harder supplier onboarding |
What is Point cloud?
Point cloud should be requested when the buyer's training or evaluation pipeline already expects 3D scene geometry, object reconstruction, LiDAR/depth capture, navigation perception, and manipulation planning. Anchor the bounty to the canonical specification before suppliers submit samples [1], then use implementation documentation to make the expected file layout reviewable [2]. Robotics teams should also name the dataset or paper lineage they expect suppliers to support [3].
[1]"The Point Cloud Library (PCL) is a standalone, large scale, open project for 2D/3D image and point cloud processing."
For truelabel buyers, that quote matters because it turns point cloud dataset from a generic delivery preference into a source-backed requirement the supplier can test against a sample file.
Using Point cloud with robot data
A useful Point cloud sample should prove coordinate frame, units, sensor intrinsics and extrinsics, timestamps, segmentation or object labels where available, and source RGB/depth references, plus file naming, manifest completeness, timestamp behavior, and rejected-example traceability. Include at least one workflow or converter reference so the supplier can show how the files load in practice [4], one interoperability reference for adjacent formats [5], and one comparison source for why this format is preferable to a raw folder dump [6].
Related pages
Use these to move from category-level context into specific task, dataset, format, and comparison detail.
External references and source context
- Point Cloud Library documentation
PCL is a point cloud processing project used for 2D and 3D data.
Point Cloud Library ↩ - PCD file format
PCL documents the PCD file format for point-cloud delivery.
Point Cloud Library ↩ - ASPRS LAS file format exchange activities
ASPRS LAS is a constituent file-format reference for LiDAR point-cloud exchange.
ASPRS ↩ - PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
PointNet is a deep-learning reference for point sets used in 3D classification and segmentation.
arXiv ↩ - 3D is here: Point Cloud Library (PCL)
The PCL paper introduces the Point Cloud Library for 3D data processing.
IEEE ↩ - HDF5 1.14 documentation
HDF5 can store structured arrays associated with point-cloud metadata.
The HDF Group ↩
FAQ
What is Point cloud used for?
Point cloud is used for 3D scene geometry, object reconstruction, LiDAR/depth capture, navigation perception, and manipulation planning.
What fields should Point cloud delivery require?
At minimum, require coordinate frame, units, sensor intrinsics and extrinsics, timestamps, segmentation or object labels where available, and source RGB/depth references, plus a delivery manifest and validation notes.
Can suppliers convert into this format?
Some suppliers can deliver directly in the requested format; others may need conversion. Buyers should require a small sample before full delivery.
Should the format be decided before capture?
Yes. Deciding the format before capture prevents missing fields, timestamp drift, and expensive post-delivery cleanup.
Working with point cloud dataset
Truelabel normalizes point cloud dataset across capture partners so you can ingest one consistent schema instead of writing per-vendor adapters.
Request Point cloud data