This is an old revision of the document!
Pointcloud labeling for training
This video is about labeling pointcloud to serve custom AI-training jobs and create detectors to be used in the Analysis Engine AI.
Step by step
- Open the pre-prepared Mobile Mapping Resource
- Inspect the source dataset
- The image folder should not contain processed images
- Only .jpg image format is accepted
- Download latest detectors from Context Insights detectors download page
- Launch the Analysis Engine
- Start the cracks detection
- Import as objects the resulted detected image pixels/annotations
- Convert image annotations into objects
- Inspect the resulted objects
Used Data
- The Mapping Resource used for this video: “ContextInsights cracks detection”
- The instructions to download: Notes & Sample Data for Tutorials
Reference Documentation
Last modified:: 2024/02/06 15:03