About the Reality Capture Process
While a photograph lets you capture the real world in a 2D representation, a laser scan lets you capture the real world in 3D representation. Think of it like a 3D photograph, but the typical representation is a point cloud. These clouds are comprised of millions of colored points in a precise 3D space. ReCap processes these massive data sets and provides the ability to aggregate, enhance, clean, and organize the data, and also prepares it for use in other Autodesk products.
The path from capturing raw scan or photo data, refining it and utilizing it in a CAD program has three main phases, as follows.
1. Data capture
A series of laser scans or photos provides the source data.
In a scan project, point data is captured using 3D laser scans, and aligned with a referenced coordinate system. Generally, a survey company provides correctly oriented scans in the native scan format of the scanner manufacturer.
In a photogrammetry project, often called simply a photo project, ReCap Pro analyzes a series of photos to measure distances between objects, map exact surface points, and create mesh. As a result, users can develop textured, high-resolution, geometrically precise 3D models from simple photographs. This feature leverages cloud service on the web at Autodesk Drive.
2. ReCap project creation
The project creation process differs, depending on whether you are using laser scans or photos as source data.
Scan data is imported, indexed, and saved as a ReCap project. ReCap converts the scan files to a proprietary Reality Capture Scan format (RCS) that can be read by other Autodesk programs.
At the start of this process, you can adjust the number of points imported by changing settings for noise, distance from the camera, intensity range, and decimation values. Save the indexed scan files in a Reality Capture Project file (RCP) that references, but does not contain, the files. ReCap supports the import of structured scans (including RealViews and registration where applicable) for individual or multi-scan E57 files.
Photogrammatic data consists of digital photos taken from an unmanned aerial vehicle (UAV) or a number of fixed camera positions. The key requirement is that the field of view for each photo overlaps with those of adjacent photos on all sides so that all points around the subject are captured. After you import the photos into ReCap, you can review and edit the set of images, and then start the process of Photo-to-3D stitching or registration, which creates a seamless 3D panoramic scene. Stitching can occur automatically in ReCap, but if you have ReCap Pro, you can manually refine the process by designating survey points, such as precise corners of a building, that are visible in two images.
3. Organization and analysis
The scene and its objects are measured, filtered, cleaned, and organized.
A number of methods are available to organize the indexed data to remove or hide portions of the point cloud. Create Scan Regions that you can turn on and off, specify temporary clip regions, turn off or remove scan files, or delete unwanted points permanently. By viewing the scene using different color modes and lighting, you can gain insight into elevations, normals (which help identify surfaces), and reflectivity.
Indexed scan files and projects can be opened or attached in other Autodesk programs to provide realistic context for 3D models.