Data Collection

Prev Next

Not all scan data is of the same quality and varies based on the sensor used, the environment being scanned, collection methods, and the quality of the data processing. The final visual representation, whether viewed as a point cloud or mesh, depends on the quality and completeness of the input data. Following proper capture procedures and producing well processed datasets ensures the best visual fidelity and user experience inside Prevu3D.


Collection Requirements

As Prevu3D offers a high resolution mesh solution, it is important to capture data properly with the right equipment for the best results. The following is the recommended requirements to properly capturing the environment data.

Terrestrial Laser Scanning (TLS)

Terrestrial scanning, also known as terrestrial laser scanning (TLS) or ground-based lidar scanning, is a technology used to capture highly detailed and accurate three-dimensional (3D) data of objects or environments.

SLAM Scanning

SLAM, which stands for Simultaneous Localization and Mapping, is a technology used in robotics and computer vision to enable a device, such as a robot or a camera, to map its environment while simultaneously determining its own position within that environment. SLAM works by using various sensors, such as cameras, lidar (light detection and ranging), and IMU (inertial measurement unit), to collect data about the surroundings.

A simplified distinction: TLS is static, SLAM is moving while scanning.

SLAM scanners are often more cost-effective due to their ease of use and mobility. They are favored when timely data acquisition is a priority. That being said, speed will have an impact on the quality. Terrestrial scanners will take more time, but will achieve higher quality if properly done.

Drone Photogrammetry

Drone photogrammetry captures structures, sites, and landscapes using overlapping aerial images. These images are processed to generate point clouds, meshes, and textured 3D models.

Drone photogrammetry is ideal for:

  • Outdoor environments, large sites, roofs, facades, and inaccessible areas

  • Supplementing TLS/SLAM to cover areas where terrestrial scanning is not feasible

  • Capturing large-scale geometry efficiently

While drone photogrammetry produces visually rich meshes and wide coverage, its geometric accuracy depends heavily on flight planning and the quality of image alignment. For the best results, consider combining drone data with TLS or SLAM scans.

360° Video Photogrammetry

360° cameras can be used to capture environments using video-based photogrammetry. Instead of capturing static scans or images, a 360° video is recorded while walking through the environment. This approach provides a simple and cost-effective entry point into the Prevu3D ecosystem, allowing users to start documenting spaces using lightweight and widely available equipment.

Typical use cases include:

  • Quickly documenting indoor facilities

  • Capturing environments with minimal setup or equipment

  • Creating visual documentation of spaces for inspection, planning, or collaboration

  • Providing an accessible first step before adopting higher-precision capture technologies

Many organizations begin with 360° capture workflows due to the low hardware cost and fast capture process, and later expand to higher-end technologies such as SLAM or TLS scanning when higher geometric accuracy becomes necessary.

While the processing pipeline generates a 3D mesh representation, geometric accuracy should not be considered the primary objective of this capture method. The mesh mainly serves as a visual container for spatial navigation.

The main value of 360° video capture is the ability to:

  • Navigate environments through immersive photosphere views

  • Visually explore spaces and equipment

This workflow provides strong visual context for environments and lays the foundation for future visualization technologies, including Gaussian Splat-based rendering.

To ensure successful processing, proper capture techniques and camera settings must be followed.

For full capture instructions, see:
360° Camera Capture Guidelines

Exporting captured scans

Before uploading data, it is necessary to export the captured data depending on the capture device and software used. The better the scanner and quality of the registration, the better the output quality and accuracy is.

Depending on the supported capture device used for scanning, the export settings will need to be curated to satisfy all requirements for the upload process.

For the highest quality that Prevu3D has to offer, it is recommended to use high-end TLS scanner like Leica RTC360, FARO S Series, etc.. At the end of the day, the quality required really depends on our needs and should be communicated clearly while scoping the scanning project.

Scan completeness

To avoid an incomplete result and limit occlusion of the environment, it is necessary to scan an environment thoroughly to ensure the capturing of all details.

  • As a general practice, scan setups should be spaced 3-5 meters (10 to 16 feet) apart.

  • Additional scans around equipment, objects of importance, and in congested areas are recommended to ensure no missing details.

Scan lighting

Scan an environment with sufficient lighting and visibility to ensure detailed capturing.

  • Proper lighting:

    • Ensure there is quality lighting in the space you are scanning

    • Dark areas will need to be lightened to correctly record the data

    • Avoid overexposed lighting from windows or spotlights

  • Color balancing of imagery

  • HDR imagery preferred

Static environment

To avoid distorted errors within the captured data, it is recommended to scan the environment at a time where there is little to no movement nor operations being performed within the target view of interest

  • Recommendations:

    • Scan during off-peak hours

    • Eliminate / limit people and equipment moving in scans

    • If completing multiple days of capture, try to capture complete areas on the same day to minimize potential movement of objects between days of capture

Best practices

We have built 2 Whitepapers related to Best scanning practices:

Info

Read more about supported data here.