Automated scan-to-CAD for as-built documentation, planning, and renovation
3D laser scanning and photogrammetry capture building facades quickly, accurately, and in high detail. Our AI analyzes the resulting point cloud, identifies relevant facade elements, and generates CAD- and BIM-ready geometry from it – automated, traceable, and dimensionally accurate.
Facade evaluation in progress.
We are developing it together with partner architecture firms using real project data. Would you like to contribute your facade scans? Talk to us about a pilot project.
Scan-to-CAD Instead of Manual Tracing
The point cloud captures the actual geometry of a building in great detail – making it an excellent measurement basis. However, further planning requires clearly defined geometric elements, not raw measurement points.
Today, these structures are often traced manually from the point cloud. For larger buildings or many objects, this creates considerable effort. Our AI automates exactly this work step.
Required Geometric Elements
- Outer contours and facade axes
- Building corners and window openings
- Door openings and heights
- Sill and lintel lines
- Floors and roof edges
- Balconies, projections, and recesses
Input
3D point cloud from laser scan or photogrammetry
AI Processing
Automatic recognition of facade elements
Output
DXF as standard; DWG, IFC, and GIS formats on request
Automated Building Envelope Analysis
Building facades consist of a combination of different geometric structures. Large wall surfaces are often approximately planar, windows and doors form openings, balconies and cornices protrude from the plane, and loggias and recesses lie behind it.
Planar Surfaces
Large wall areas are segmented using normal vectors and flatness analysis.
Openings
Windows and doors are detected as geometric interruptions in the facade plane.
Protrusions
Balconies, cornices, and loggias are delineated using the depth information of the point cloud.
Edges
Building corners, roof edges, and parapets emerge where the surface orientation changes.
Our solution combines machine learning with classical 3D geometry analysis to reliably process both types of structures.
Recognizing Facade Surfaces
First, contiguous wall areas are identified. The analysis considers normal vectors, flatness, orientation, point density, spatial continuity, and depth structure. From millions of individual measurement points, defined facade planes are created.
Multiple facades of a building can be automatically separated from one another. Each surface receives a complete geometric description.
Properties per Facade Surface
Building Edge Detection
Building edges form where the spatial orientation of a surface changes or where a facade plane ends. The AI reliably detects these discontinuities and converts them into geometrically stable CAD lines.
Building Corners
Vertical edges at facade intersections
Roof Edges & Parapets
Upper building termination, horizontal
Base Edges
Transitions to ground level
Cornices
Horizontal facade profiling
Floor Transitions
Lines between story zones
Projections & Recesses
Depth-offset facade areas
Windows and Doors
Analysis Parameters
- Geometric interruptions
- Depth changes in the facade
- Vertical and horizontal edges
- Symmetries and repetitions
- Facade grid
- Image data (where available)
Window and door openings are especially relevant for as-built documentation and planning. The AI analyzes geometric interruptions, depth changes, and edge patterns. Where image data is available, it is included as well.
Output Parameters per Opening
Balconies, Loggias, and Projections
Balconies and loggias are three-dimensional structures that stand out spatially from the main facade plane. The depth information in the point cloud enables automatic delineation of these elements. The detected geometries are spatially assigned to the respective facade surface.
Balcony Slabs and Parapets
Horizontal and vertical delineation of the balcony structure using edge geometry
Loggias and Recesses
Recessed facade areas behind the main plane, with depth determination
Facade Offsets
Local projections and recesses within the facade surface, relevant for prefabrication and connection details
Roof Edge and Parapet
The upper building termination is often especially important for facade, roof, and renovation planning. The AI detects this transition zone with high geometric precision and outputs the results as defined lines or 3D geometries.
Roof Edge
Upper termination edge of the facade at the roof area, as a dimensionally accurate 3D line
Parapet
Parapet-like masonry rising above the roof edge, with front and top edges
Cornice
Profiled facade termination between the facade and the roof
Upper Facade Boundary
Geometrically defined transition between the facade surface and the roof membrane
From Scan to CAD Model
The transformation from point cloud to usable CAD geometry follows a structured process with several sequential processing steps. Each step builds on the previous one, enabling controlled and traceable processing.
2D Facade Elevations
Not every project requires a full 3D or BIM model. For many applications, a dimensionally accurate 2D facade elevation is the most efficient deliverable. The point cloud is projected onto a defined facade plane, the relevant geometry is extracted, and delivered as DXF (DWG on request).
All elements can be output on separate CAD layers, so the drawing can be directly further processed in existing CAD software.
Typical Drawing Content
- Exterior outline and building corners
- Ground line and plinth
- Windows and doors
- Floor lines
- Roof edge and parapet
- Balconies and loggias
- Cornices and facade projections
- Recesses
3D CAD and Scan-to-BIM
In addition to classic 2D facade elevations, the detected geometry can also be built up in three dimensions. Planar facade surfaces are positioned in space, and windows, doors, and other elements are assigned to their respective surfaces. Depending on the application, different levels of modeling detail are possible.
Basic Geometric Model
Building envelope, main facades, essential openings
Detailed Facade Model
Windows, doors, balconies, loggias, cornices, facade grids
BIM Foundation
Structured BIM objects for further planning processes
Note
IFC output available on request.
Modeling Depths
Basic Geometric Model
Included Elements
- Building envelope
- Main facades
- Building corners
- Roof edge
- Essential openings
Suitable For
- Existing condition survey
- Early planning phases
- Area calculation
- Conceptual design
Detailed Facade Model
Additionally Modeled Elements
- Windows and doors
- Balconies and loggias
- Parapets and plinths
- Cornices
- Facade grid
Suitable For
- Renovation planning
- Facade planning
- Tendering
- Prefabrication
Basis for BIM (on Request)
Depending on the project, the detected geometries can subsequently be assigned to structured BIM objects. This creates a reliable geometric foundation for further BIM processes.
The required modeling depth is defined before the project begins. As a result, not every detail is unnecessarily modeled — only the information actually needed for the specific use case is generated, which targets effort and costs precisely.
Deviation Analysis (On Request)
An existing building facade is rarely perfectly flat, vertical, or right-angled. Our solution compares the actual surface with a geometric reference and quantifies deviations precisely.
Flatness
Distances between measurement points and the reference plane – shown as a numerical value or color map
Plumb Deviation
Systematic inclination of the facade relative to the ideal vertical
Offsets and Recesses
Distances of the surface from a planned reference plane for new facade systems
Serial Retrofit and Prefabrication
Scan-to-CAD is especially relevant for projects where facade elements are manufactured industrially. Production requires reliable information about the as-built condition – not the planned target state.
Survey
Laser scan of the existing building stock
Planning
CAD geometries and deviation analysis
Prefabrication
Dimensionally accurate production of facade elements
Installation
Precision-fit installation on site
Our AI derives facade width, window positions, story heights, connection geometries, and local irregularities directly from the point cloud and provides them in a structured form for the next process step.
Occlusions and Data Sources
Occlusions
Point clouds only contain the areas that were actually captured. Trees, vehicles, balconies, scaffolding, as well as reflective or transparent surfaces, cause gaps and problematic measurements.
Such areas are not adopted automatically but receive a review status for targeted post-processing.
Data Sources
Combining terrestrial laser scanning, mobile mapping, drone LiDAR, and photogrammetry reduces occlusions and provides additional information for detection.
Quality Assurance
Quality assurance is just as critical for building facades. Several quality attributes can be calculated for each detected element. This keeps the AI output traceable and controlled, allowing uncertain areas to be reviewed and refined in a targeted way.
An object class only goes into production once it reaches at least 90% AI maturity – the share of element length that can be accepted without rework – and meets the agreed quality criteria.
Human in the Loop
The AI handles repetitive analysis and digitization work. Expert review remains part of the process. This approach combines the scalability of automated processing with the reliability of human expertise.
Analyze the point cloud
Facade surfaces, edges, and building components are detected automatically
Generate CAD geometry
Point groups are converted into lines, surfaces, and structured objects
Assess quality
The AI evaluates reliability and geometric plausibility
Flag critical areas
Occlusions, unusual geometries, or weakly captured areas are highlighted
Expert sign-off
Critical areas are reviewed in a controlled manner, and the final CAD or BIM output is approved
Standard – DXF (2D Facade Elevations and 3D Polylines)
On Request – DWG, IFC, GeoJSON, SHP, GeoPackage, LAS/LAZ/E57, QA Data
Standard: DXF (DWG on Request)
Dimensionally accurate facade elevations with a defined layer structure – the most commonly used output format for direct use in existing planning workflows. All elements are delivered on separate, named layers.
The format is directly compatible with the most common CAD systems and requires no conversion or preprocessing on the recipient's side.
Compatible Systems
- AutoCAD
- BricsCAD
- ARCHICAD
- Existing planning workflows
- As-built plan workflows
3D CAD
3D polylines are available; further 3D outputs on request. The detected facade geometry serves as a direct basis for further design, clash detection, or visualization. The spatial positioning of all elements is fully preserved.
Lines and Polylines
Edges and axes in 3D space
Planar Surfaces
Facade planes with correct spatial position
Openings
Windows and doors as 3D objects
Building Elements
Balconies, cornices, and parapets spatially assigned
On Request: BIM / IFC
Depending on the project, the detected geometries can be prepared for structured BIM workflows. Output as IFC enables direct import into BIM authoring tools such as Revit, ARCHICAD, or Allplan.
The required level of detail is defined before the project begins. This ensures that the BIM output matches actual project requirements without unnecessary modeling effort for details that are not needed.
On Request: GeoJSON / SHP / GeoPackage
For georeferenced building information and connection with GIS systems, open vector formats can be provided on request. The detected facade geometries can be output with coordinate reference system and attribute data and imported directly into QGIS, ArcGIS, or comparable systems.
This format is particularly relevant for property management, municipal planning processes, and portfolio documentation across multiple buildings.
On Request: LAS / LAZ / E57
The point cloud is retained as a geometric reference. On request, a classified point cloud can additionally be provided, in which each point is assigned a semantic class – such as wall, window, balcony, vegetation, or ground.
The classified point cloud enables visual quality control and can serve as a basis for further automated processing steps.
On Request: QA Data
Optionally, quality information can be delivered together with the geometric result. The QA data enables targeted post-processing and documents the reliability level of each detected element.
Coverage Information
Share of captured vs. uncaptured facade areas
Review Zones
Marked areas with limited detection quality
Reliability
Confidence value of the automatic classification per object
Deviation Values
Numerical distances between CAD geometry and measured points per element
Model Version
Versioning for traceable result history
Typical Applications
Facade Renovation
Current, dimensionally accurate basis for planning on the existing building
Revitalization and Conversion
Digital as-built survey when existing plans are missing or deviate from the actual building condition
Curtain Wall Facades
Precise geometry as the basis for new facade systems and substructures
Serial Retrofit
Digital measurement for industrially prefabricated facade and retrofit elements
Building Documentation
Measurable digital condition of a building for facility management and portfolio analysis
Scan-to-BIM
Geometric foundation for structured BIM workflows according to DIN EN ISO 19650
Heritage Conservation
Documentation of complex historic facades and building components with high level of detail
Cloud as the Standard – On-Premise When Needed
Our AI solution is delivered as a cloud service by default. This allows large point cloud processing to scale flexibly without requiring the customer to build additional GPU infrastructure.
For projects with particularly high security or data privacy requirements, all processing can also be run on-premise within the customer's own IT infrastructure.
The standard is the Jaroona Cloud. Private cloud, customer-owned cloud, on-premise, and hybrid models are available on request.
Supported Deployment Models
- Jaroona Cloud
- Private Cloud
- Customer-Owned Cloud Environment
- Fully On-Premise
- Hybrid Architecture
Strict Data Privacy and Data Sovereignty
Point clouds of buildings can contain detailed information about architecture, construction, access points, and technical installations. Protecting this information is a central part of our solution.
Storage Location
The client determines where data is stored and how long it remains available
Access
Who receives access and which results may be exported remains fully controllable
Processing Location
With on-premise deployment, all data remains within the client's own environment
Deletion
Defined retention periods and controlled data deletion after project completion
Processing takes place on servers in the EU. Data transmission is encrypted, project workspaces are separated, and access rights are role-based.
Test Your Own Point Cloud
Provide us with a representative section of a building point cloud and the desired CAD structure. Together we will review which facade elements can be detected automatically, what data quality is required, and which output formats integrate best into your existing processes.
Detectable Elements
Which facade components can be captured automatically?
Data Quality
What minimum requirements apply for point density and scan coverage?
Achievable Accuracy
What level of detail and dimensional accuracy are realistic for your project?
Process Integration
How do the results fit into your existing CAD or BIM workflows?
Survey2CAD
AI that turns point clouds into finished CAD plans.
© 2026 Jaroona. All rights reserved.
Point Cloud to CAD for Building Facades