Knowledge Base - Version 22.4
Recent Changes

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
Next revision Both sides next revision
224:desktop_ext:context_insights [2022/06/13 14:25]
127.0.0.1 external edit 		224:desktop_ext:context_insights [2022/06/22 15:52]
jeroen
Line 1: Line 1:
 ====== ContextInsights ====== ====== ContextInsights ======
  
-This page describes how to use the extension "ContextInsights"\\+This page applies to the following products: 
  
-{{orbit_desktop:how_to_find.png?25&nolink  |}} [[224:desktop:workspace:main_toolbar|Main Toolbar]] > Extensions > Context Insights+  * Orbit 3DM Feature Extraction Pro – from 22.04 - CONNECT Edition 
  
-The "ContextInsights" extension makes it possible to use Context Insights detectors on mapping resources and pointcloud resources. The detectors are Image 2D Annotations and 2D Segmentations.  
  
 +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 [[https://kb.orbitgt.com/dev/technology/platforms/install_contextinsights#contextinsights_-_installation|dedicated page]]\\
  
-===== Concepts ===== +{{orbit_desktop:how_to_find.png?25&nolink  |}} [[224:desktop:workspace:main_toolbar|Main Toolbar]] > Extensions > Context Insights
- +
-Running Context Insights detectors requires Python version 3.6, ContextCapture Center Engine and original images from any mapping resource. The results can be directly imported in the mapping resources via the Annotations procedure. Install the [[224:technology:platforms:install_contextinsights|Context Insights Dependency]] before running Context Insights jobs.  +
- +
-===== Preferences ===== +
- +
-Configure the directories to Python, the virtual environment and the Context Insights detectors. +
-The detectors and specs are listed underneath.  +
- +
-===== Action ===== +
- +
-Create a new job and enter the target location of the results.  +
- +
-Open an existing job and browse to the location of the results. +
  
-Drag and drop a directory from file explorer into Orbit. +ContextInsights extension enables automatic detection on mapping and pointcloud resources.\\ 
 +It relies on detectors to identify objects or regions of interest.
  
-===== Annotation Type ===== 
  
-Depending on the added detectors, the following annotation types will be selectable.+===== Jobs description =====
  
-==== Image 2D Objects ====+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 [[https://communities.bentley.com/products/3d_imaging_and_point_cloud_software/w/wiki/54656/context-insights-detectors-download-page|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
  
-  * Import the scene xml file as 2D Object annotations to overlay the results on original or optimized images+===== Jobs execution =====
  
-==== Image 2D Segmentation ====+In this section, it is assumed ContextInsights has been installed as recommended and Preferences have been set.\\ 
 +\\
  
-  * Apply Segmentation at optimize imagery to display the result on optimized images+  - Launch ContextInsights Engine\\ {{:224:desktop_ext:cinsights_engine.jpg?200|}} 
 +  - Set in Extensions 
 +  - ContextInsights\\ {{:224:desktop_ext:extensions.jpg?200|}} 
 +  - Define whether you want to create a new job or open existing one 
 +  - Choose the output folder of the detection 
 +  - Choose annotation type:\\ **IMPORTANT**: Image based jobs can only run on JPG images\\ {{:224:desktop_ext:annotations_type.jpg?200|}} 
 +      * 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\\ {{:224:desktop_ext:detector.jpg?200|}} 
 +  - Define the source data detection job must occur on 
 +  - Start job\\ {{:224:desktop_ext:start.jpg?200|}}
  
-==== Pointcloud 3D Segmentation ==== 
  
-===== Analyze ===== 
  
-Delegate the process to the task manager or start now. Running the Context Capture Engine is required before starting the process.  
  
 ===== Import ===== ===== Import =====
 
Last modified:: 2022/06/23 09:38