The world is composed of two types of information: digital and physical. Physical representations of digital data exist all around us, as do digital representations of physical data. You can look through billions of digitized photos on the internet if you have a computer and network connection. You can look through vast libraries of 3D models that represent physical objects. There are physical items that you possess made from digital designs modeled after physical objects.
Of course, in the physical world of information, we have many manmade structures, like buildings and bridges, that were designed and built before the information technology age, during and prior to the industrial revolution. The surface geometry of these structures can now be recorded or captured from the physical world and processed into digital data. This digital data can be manipulated by powerful software for a variety of practical reasons in architecture, engineering and construction (AEC) and other sectors.
In fact, this capability is empowering architects, engineers and construction teams with the ability to incorporate as-built 3D models into their planning and diagnosis of ongoing operations. The ability to capture reality data with 3D scanning hardware and use as-built 3D models is changing the way some traditional AEC workflows are approached. This approach is becoming known as “reality capture”.
What Is Reality Capture?
With photogrammetry software, users can stitch together their photos manually or automatically, and can even leverage GPS data as reference coordinates to improve the quality of the final 3D model.
Laser scanners collect surface data about an object or structure by beaming a laser repeatedly off its surface and plotting points (equivalent to each beam that is fired) into what’s known as a point cloud. Once this surface topology data is collected, it is formatted in the native file type of the laser scanner being used.
After this, the point cloud data can be imported into photogrammetry software, such as Autodesk ReCap. ReCap allows users to create an accurate 3D model within a few minutes, and has some time-saving features like automatically removing common but unnecessary analogous objects, such as people, vehicles, etc.
Autodesk ReCap can use captured data from a variety of sources, but is primarily implemented to process laser scan data and digital photographical data to create as-built 3D models that can be used in a digital assembly line of other software, including Revit and Infraworks.
Examples of Reality Capture
For one project, Turner Fleischer Architects Inc. invested in a suite of reality capture technology and was able to reduce rework, collect and use accurate as-built models of typically inaccessible spaces, increase the quality of collaboration between project teams, improve the budgeting process and realize a 325 percent return on investment (ROI).
The team that was assigned to convert 12,000 square feet of office space and an underground parking garage into an upscale food store, which required renovating two levels of office space, including increasing the minimum headroom space from 13 feet to 22 to 24 feet on the ground level, where the store was to be located. Beneath the ground level was an underground parking garage that had to be remodeled to house a compressor room for the store.
For Mercy Hospital in Jefferson, Mo., designers from Bates Architects were charged with the task of renovating and adding new entrances to new main hallways, new clinics and other facilities that don’t currently exist for patients. By using laser scanning to capture the as-built data of the hospital, Bates Architects recorded a significant decrease in project lifecycle times for a large-scale project.
Australia-based Atlas Engineering used reality capture technology to increase stakeholder buy-ins through the creation of dynamic visualizations and flyovers created from point clouds and 3D models captured by laser scanners.
A crew from Atlas Engineering’s Water Group partnered with Sydney Water, Lendlease and John Holland to engineer a massive program of modifications, improvements and new works for Australia’s pumping stations, treatments plants, reservoirs and waste water networks, as well as its largest water networks.
They decided to use a FARO Focus3D X 330 camera and scanned 60 separate projects. The 3D data would consist of over 50 laser scans, and they used Autodesk ReCap 360 to turn the scan data into 3D models and 2D drawings, which could then be shared with different collaborators.
Digital photogrammetry is also being used by inventive young scientists like Sly Lee of the Hydrous Project to capture photogrammetric data that is processed by powerful software to stitch the photos together and make a 3D model out of sets of photos of an object. Lee used these new techniques to capture the as-built 3D data of coral reefs. His project continues to do this now, and it serves as an example of how digital photogrammetry has enabled the 3D modeling and measuring of objects that previous technology only allowed us to estimate from 2D photography and older metrological methods.
How does Autodesk ReCap work
First, you capture point data by 3D laser scanning the structures and objects you need to capture. The scan data will be captured in the file format native to the company that produces your laser scanner, like Leica or Faro for instance.
Second, ReCap changes the files to the proprietary Reality Capture Scan (RCS) 3D file format. RCS files can be read by other Autodesk software. Also known as indexing, the raw scan files can be imported with a variety of settings the user can alter, including the number of points imported. By changing settings like decimation values, distance from camera and intensity range, a user will change the total number of points that are imported. The scan files are referenced by a Reality Capture Project (RCP) file, though it doesn’t contain any of the scan files themselves.
Users can organize indexed 3D scan data to highlight, hide or remove sections of the point cloud. They can delete individual points permanently and designate scan regions that can turn on and off, which comes in handy when users are trying to make the best version of the 3D model from the captured scan data. Users can choose from distinct color and lighting modes to identify different surfaces, elevations, reflectivity and other characteristics that help them create the best definitive version of their 3D model.
Users can then attach indexed scan files and projects in other programs to give the 3D models more realistic context.
To learn more about Reality Capture and Autodesk ReCap, click here.
Autodesk has sponsored this post. They have no editorial input to this post. All opinions are mine. —Andrew Wheeler