Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
224:desktop_ext:context_insights [2022/06/22 14:12] arnaud [Jobs] |
224:desktop_ext:context_insights [2022/06/23 09:38] (current) arnaud |
||
---|---|---|---|
Line 1: | Line 1: | ||
- | ====== ContextInsights | + | ====== ContextInsights ====== |
- | This page applies | + | This page describes how to use the " |
+ | Specific system requirements and installation guidelines, see [[224:technology: | ||
- | * Orbit 3DM Feature Extraction Pro – from 22.04 - CONNECT Edition | + | {{orbit_desktop: |
+ | ===== Concepts ===== | ||
- | It page describes how to use the ContextInsights extension\\ | + | ==== Context Insights |
- | This extension requires a specific installation procedure\\ | + | |
- | Before going any further, please make sure you reviewed the [[https:// | + | |
- | + | ||
- | {{orbit_desktop: | + | |
- | ContextInsights extension enables automatic detection on mapping and pointcloud | + | ContextInsights extension enables automatic detection on mapping and point cloud resources. \\ |
It relies on detectors to identify objects or regions of interest. | It relies on detectors to identify objects or regions of interest. | ||
+ | Running Context Insights detectors requires Python, Context Insights Engine, and original images from any mapping resource. The results can be directly imported in the mapping resources via the Annotations procedure. | ||
- | ===== Jobs ===== | + | ==== Jobs ==== |
+ | |||
+ | A ContextInsights job will leverage machine learning to execute different types of detection in your reality data. \\ | ||
+ | There are various types of jobs available: | ||
+ | * Image 2D Objects: Will detect elements as regular 2D-bounding boxes in images. Regular 2D boxes around assets. | ||
+ | * Image 2D Segmentation: | ||
+ | * Point Cloud 3D Segmentation: | ||
- | A ContextInsights job will leverage machine learning to execute different types of detection in reality data.\\ | + | Each job type requires |
- | There are various types of jobs available in Orbit Feature Extraction Pro:\\ | + | |
- | \\ | + | |
- | Detectors can be [[https:// | + | |
- | \\ | + | |
- | • Image 2D Objects: Will detect elements as regular 2D-bouding boxes in images\\ | + | |
- | • Image 2D Segmentation: | + | |
- | • 3D segmentation: | + | |
- | \\ | + | |
- | 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 | + | |
- | ===== Preferences ===== | + | ==== Detectors |
- | Configure the directories to Python, the virtual environment and the Context Insights detectors. | + | A detector is a frozen model that is pre-trained by Bentley Systems. Running on your reality data through a ContextInsights job, it will automatically recognize elements of interest. List of pre-trained |
- | The detectors | + | https:// |
- | ===== Action ===== | + | A detector is specific to a certain job type. The quality of the detection will depend on the similarity between your dataset and the training dataset’s description. |
- | Create a new job and enter the target location of the results. | + | ===== Launch Context Insights Engine ===== |
- | Open an existing job and browse to the location of the results. | + | Launch ContextInsights Engine \\ |
+ | {{: | ||
- | Drag and drop a directory from file explorer into Orbit. | + | ===== Context Insights Sidebar ===== |
- | ===== Annotation Type ===== | + | ==== Preferences |
- | Depending on the added detectors, the following annotation types will be selectable. | + | Configure |
+ | The detectors and specs are listed underneath. | ||
- | ==== Image 2D Objects | + | ==== Action |
- | | + | Define whether you want to create a new job or open existing one |
+ | | ||
+ | * Open an existing job and browse | ||
- | ==== Image 2D Segmentation | + | ==== Detectors |
- | * Apply Segmentation at optimize imagery to display the result | + | Depending |
- | ==== Pointcloud 3D Segmentation ==== | + | <note important> |
- | ===== Analyze | + | ==== Analyze ==== |
Delegate the process to the task manager or start now. Running the Context Capture Engine is required before starting the process. | Delegate the process to the task manager or start now. Running the Context Capture Engine is required before starting the process. | ||
- | ===== Import | + | ==== Import ==== |
- | ==== Open Procedure ==== | + | === Open Procedure === |
- | + | ||
- | Open the Annotations procedure where the result xml file is already pre-filled as source object annotation file. | + | |
+ | Open the Annotations procedure where the result xml file is already pre-filled as source object annotation file. \\ | ||
Import 2D Objects or 2D Segmentations, | Import 2D Objects or 2D Segmentations, | ||
- | ==== Open File Location | + | === Open File Location === |
Open the target directory to verify the results. | Open the target directory to verify the results. | ||
+ | |||
+ | === Image 2D Objects === | ||
+ | |||
+ | Import the scene xml file as 2D Object annotations to overlay the results on original or optimized images | ||
+ | |||
+ | === Image 2D Segmentation === | ||
+ | |||
+ | Apply Segmentation at optimize imagery to display the result on optimized images | ||
Last modified:: 2022/06/22 14:12