For documentation on the current version, please check Knowledge Base.
This is an old revision of the document!
ContextInsights - Execution
This page applies to the following products:
- Orbit 3DM Feature Extraction Pro – from 22.04 - CONNECT Edition
It page describes how to use the ContextInsights extension
This extension requires a specific installation procedure
Before going any further, please make sure you reviewed the dedicated page
Main Toolbar > Extensions > Context Insights
ContextInsights extension enables automatic detection on mapping and pointcloud resources.
It relies on detectors to identify objects or regions of interest.
Jobs description
A ContextInsights job will leverage machine learning to execute different types of detection in reality data.
There are various types of jobs available in Orbit Feature Extraction Pro:
Detectors can be downloaded here before being unzipped and used by your instance
• Image 2D Objects: Will detect elements as regular 2D-bouding boxes in images
• Image 2D Segmentation: Will detect regions in images and deliver segmented raster
• 3D segmentation: Attribute a class to each point of laserscan data
Each job type requires suited detector to be executed on reality data, e.g: a 2D segmentation detector will only be usable for image based segmentation jobs
Jobs execution
In this section, it is assumed ContextInsights has been installed as recommended and Preferences have been set.
- Launch ContextInsights Engine
- Set in Extensions
- ContextInsights
- Define whether you want to create a new job or open existing one
- Choose the output folder of the detection
- Choose annotation type:
- 2D objects: Regular 2D boxes around assets
- Image 2D segmentation: Regions to segmented raster
- 3D segmentation: classify each element of a pointcloud
- Choose a detector depending on the information you aim at extracting
- Define the source data detection job must occur on
- Start job
Import
Open Procedure
Open the Annotations procedure where the result xml file is already pre-filled as source object annotation file.
Import 2D Objects or 2D Segmentations, depending on the chosen Annotation Type.
Open File Location
Open the target directory to verify the results.