10 Things You Can Do with a Point Cloud Viewer｜Explained from an On-site Perspective: Measurements, Cross-sections, Annotations, and Comparisons

Table of Contents

• What You Can Do 1. Intuitive 3D understanding by viewing point cloud data

• What You Can Do 2. Quantification by measuring distance, area, and volume

• What You Can Do 3. Precise shape inspection through extraction and display of arbitrary cross-sections

• What You Can Do 4. Adding and sharing information through annotations and tagging

• What You Can Do 5. Verification by overlaying with other data

• What You Can Do 6. Monitoring by comparing point cloud data and extracting changes

• What You Can Do 7. Detailed analysis by checking point attribute information

• What You Can Do 8. Data consistency through coordinate transformation and unification of reference systems

• What You Can Do 9. Efficient operation through data reduction and optimization

• What You Can Do 10. Smooth communication through sharing and collaboration on point cloud data

• Summary

In recent years, the use of point cloud data captured by 3D laser scanners and photogrammetry has rapidly spread across construction, civil engineering, and surveying sites. However, it has long been a major challenge to easily display and share three-dimensional data composed of tens of millions to hundreds of millions of points. Dedicated software installation and high-performance PCs were required, making it difficult to casually check 3D data on site or share it with stakeholders. A solution that has attracted attention for lowering this barrier is general-purpose point cloud viewers—application-type viewers that run on Windows PCs and cloud-based viewers that run in web browsers. By using viewers that support the LAS and LAZ formats, even massive point clouds can be viewed smoothly, and a wide range of functions from measurement and analysis to data sharing can be utilized on site. With cloud-based viewers, no dedicated software is necessary and they can be used from tablets and smartphones; by uploading data and sending a shared link to stakeholders, they can view the latest 3D point clouds, making internal and external information sharing dramatically easier. In this article, from the perspective of practitioners, we explain concrete usage methods for ten representative functions provided by general-purpose point cloud viewers that support LAS/LAZ formats.

What You Can Do 1. Intuitive 3D Understanding by Viewing Point Cloud Data

The core function of a point cloud viewer is, above all, the ability to display and browse 3D point cloud data. The viewer visualizes three-dimensional data made up of countless points, allowing users to freely rotate and zoom to observe it from any angle. This enables an overview and inspection of the entire site as if you were there, and lets you intuitively grasp height information and complex terrain shapes that were difficult to capture in traditional flat drawings. For example, you can check the contours of a planned development site from above or examine the detailed shape of a structure from various angles—point cloud data functions as a “digital copy” of the site. Parts that were unclear in 2D drawings or photographs become immediately evident when viewing a 3D point cloud, enabling all stakeholders to share a common spatial understanding.

Especially for point clouds of vast terrain, they can accurately visualize subtle surface undulations that were difficult to grasp from maps and aerial photographs alone, helping to correctly understand the topography and make decisions during the design and construction planning stages. Steep slopes and high locations that are dangerous and inaccessible on site can be safely checked on the point cloud, and checks from a bird’s-eye viewpoint that allow you to virtually walk around the site also help prevent overlooking risk factors.

What it can do 2. Quantification by measuring distance, area, and volume

The point cloud viewer is equipped with measurement tools that can instantly calculate various dimensions and quantities from acquired point clouds. For example, specifying any two points will give the distance, and enclosing an area with a polyline drawn on the spot will automatically measure the area.

This allows you to grasp distances and extents on the point cloud data without using tape measures or surveying instruments on site. For example, in applications such as measuring the width of an excavation or calculating the paved area of a site, the site supervisor can immediately confirm the required dimensions.

Also, in recent years viewers have enhanced their volume (earthwork) calculation functions. By scanning mounds of fill or spoil with point cloud data and simply enclosing their outlines, fill and cut volumes can be calculated automatically. On the viewer, if you select the target area with a polygon and designate the surrounding ground height as the reference surface, the amount protruding from the reference surface can be quantified instantly. For example, you can grasp the approximate volume of fill created by heavy machinery or the soil quantity in an excavation trench on site, enabling you to perform earthwork volume calculations instantly that used to be done back at the office. These measurement functions drastically improve the efficiency of the quantity checks required for as-built management. In fact, at one large-scale site, surveying and calculating as-built earthwork volumes that had taken 4 people × 1 week (a total of 28 person-days) was switched to drone point clouds and viewer-based volume calculation and completed in 2 people × 1 day (2 person-days).

What You Can Do 3. Precise Shape Verification by Extracting and Displaying Arbitrary Cross-Sections

In point cloud viewers, it is common to have a function that slices an object at arbitrary cross-sections and displays them. Rather than merely viewing vast point cloud data from above, slicing at required locations to view cross-sectional shapes allows detailed analysis of internal structures and elevation differences. For example, displaying a longitudinal section along the centerline of a road or embankment enables accurate understanding of longitudinal gradients and elevation changes. In the case of tunnels and bridges, extracting cross-sections at arbitrary positions makes it easy to compare with design sections and to check clearances (headroom).

By dragging the cross-section plane in the viewer, you can sequentially check continuous terrain changes and the shapes of structures, gaining a three-dimensional understanding that paper cross-sections could not provide. In particular, for ground surveying, you can generate longitudinal and cross-sectional drawings from surface point clouds at any location, making it possible to create design cross-section materials without relying on additional field surveys. These cross-section extraction functions are also useful for as-built inspections of civil engineering structures. By displaying designated measurement sections from point cloud data of the ground and structures after construction and checking whether their shapes conform to the design values, you can detect construction errors early and improve the efficiency of quality inspections. Tasks that traditionally required placing survey points on site for each section can, with only point cloud data, be analyzed retroactively for any section, dramatically increasing the flexibility of surveying methods.

What You Can Do 4. Adding and Sharing Information via Annotations and Tagging

A point cloud viewer not only lets you simply view data, but also enables you to add information by annotating (comments) and tagging the data. You can write important items or measurement results as text directly onto the site’s point cloud, or place markers at specific locations so they can be referenced later. For example, if you leave a note such as "the condition of location XX as of month/day" at the corresponding position in the point cloud, you can easily grasp the condition and annotations of that place even when you review the data after some time has passed. By attaching tags and comments, you can immediately reach the "desired information" within vast point cloud datasets, dramatically improving the searchability of field records.

In practice, information that tends to get buried in photos or paper notes can be linked directly to point cloud data, allowing intuitive spatial sharing—for example, “that damaged section of piping” or “a location suspected of a construction error.” In addition, with cloud-based viewers, stakeholders can view the same annotated point cloud model online, so comments about issues found on site can be shared immediately with headquarters or partner companies, preventing misalignment in understanding. By using annotation and tagging features in this way, point clouds can be used not merely as geometric data but as “living documents” that incorporate on-site knowledge and history.

Please translate the following input into English.

What it can do 5. Cross-checking by overlay display with other data

A powerful feature of a general-purpose point cloud viewer is not only displaying standalone point cloud data but also overlaying it with other data on the viewer. A typical example is overlaying point clouds with design data or CAD drawings. For example, if you overlay an as-built point cloud after construction with the prior design model in the same coordinate space, you can visually verify whether the constructed object is in the position and shape specified by the design. In practice, comparing as-built point clouds with design data in the browser has, in some cases, enabled early detection of construction errors or missing portions that were difficult to see.

It is also possible to overlay multiple point clouds. For example, if you load point clouds acquired with different devices (such as drone surveys and terrestrial laser scanning) at the same time, you can combine aerial and ground-based data complementarily to reproduce the site in detail. Furthermore, if point clouds include geospatial coordinates (absolute coordinates such as latitude and longitude), it is also easy to display them over digital topographic maps and aerial photographs. This makes it immediately clear where the point cloud data corresponds to, facilitating position checks on existing infrastructure maps and smooth cross-referencing with other GIS information. Recently, with advances in AR (augmented reality) technology, it has become possible to composite point clouds in real time onto tablet camera images. By overlaying acquired point clouds onto the actual site scenery, you can transparently indicate the positions of buried objects not visible to the naked eye or visualize differences from design models on site. In this way, the ability to overlay other data is indispensable for relating point clouds to other elements at the site and is useful in the design, construction, and maintenance phases.

What you can do 6. Monitoring by comparing point cloud data and extracting changes

By taking advantage of the ability to handle multiple point cloud datasets, the viewer can perform comparisons between datasets and extract differences. In particular, for construction progress management and terrain change monitoring, comparing point clouds acquired at different times enables quantitative assessment of changes. For example, by overlaying the pre-construction original-ground point cloud with the post-construction as-built point cloud and calculating height differences, you can automatically compute the volumes of excavated and filled areas. Because point clouds compare high-density measurements of the entire site, high-precision earthwork volume calculations are possible, reflecting even minute irregularities that manual surveying might overlook.

Once a point cloud has been acquired, it is easy to later segment any desired area and perform additional volume calculations; for example, if part of the terrain changes due to heavy rain, you can handle it by re-calculating only the relevant area of the existing data. In addition to earthwork volumes, comparing point clouds is also effective for managing long-term deterioration. By periodically acquiring point clouds of infrastructure such as bridges and tunnels and comparing past and current data, you can capture minute displacements and deformations. Because experts do not actually have to visit the site, they can remotely assess the degree of deterioration and judge whether repairs are necessary by comparing shared point clouds, contributing to more efficient and more advanced maintenance management. In the disaster field as well, analyses are conducted such as comparing terrain point clouds before and after collapse to estimate the volume of collapsed material, or detecting deformation of structures after earthquakes.

What you can do 7. Detailed analysis by inspecting point attribute information

Each point in a point cloud dataset may have various attribute information recorded in addition to coordinates (X,Y,Z). Typical examples include the return intensity (strength) from a laser scanner and classification codes assigned to points (categories such as ground surface, vegetation, buildings). In photogrammetric and mobile-device point clouds, color information (RGB values) corresponding to each point is also added, allowing the point cloud to be displayed in color much like the real object. General-purpose viewers can color-code point clouds based on these attributes or filter (extract) points by specific attribute values. For example, switching to an elevation (height)-based color gradation display lets you intuitively grasp terrain elevation differences through a color gradient.

Similarly, when displayed by return intensity, objects that give strong laser returns—such as white lines and guardrails—stand out, making it useful for tasks like mapping road markings. If the point cloud includes classification information, it is easy to change display colors by class (ground, buildings, vegetation, etc.), hide unwanted classes, and extract and display only the objects of interest. For example, displaying only the ground surface makes terrain undulations easier to see, and displaying only trees can be used to measure forest volume. Furthermore, by clicking any point in the viewer you can check detailed numeric information for that point (coordinate values and attribute values). This allows you to read directly from the point cloud data things like the exact height at a given location or the position coordinates of an object.

What it can do 8. Data consistency through coordinate transformation and unification of reference frames

When overlaying point cloud data with other surveying results or design drawings, or comparing multiple time points, unifying the coordinate system is essential. Point clouds acquired by ground surveying or in public coordinate systems and point clouds in a drone’s geodetic coordinate system do not align as-is, so one of them needs to be transformed to match the other. Some general-purpose point cloud viewers include simple coordinate transformation functions, allowing adjustments such as specifying offset values (translation amounts) and rotation angles when loading data to match a designated local coordinate system, or converting from geodetic systems of longitude and latitude to plane rectangular coordinate systems.

When comparing point clouds before and after construction—especially when calculating earthwork volumes—it is extremely important to measure them using the same reference points and coordinate system. For drone aerial photography, place ground control points (GCPs) on site to improve absolute accuracy, and for laser scanner surveys, properly perform instrument position corrections and alignment to known points so that both point clouds overlap without discrepancy. Even if adjusting coordinates in a viewer makes them appear to overlap, if the original data’s positioning accuracy is low, the analysis will include errors. Fortunately, high-precision positioning technologies are becoming increasingly accessible. If point cloud data are assigned absolute coordinates (latitude/longitude or public coordinate system), overlaying with maps and other data and displaying in AR becomes considerably easier. Thus, coordinate transformation and unifying reference frames are important preparatory processes for utilizing point cloud data; being able to quickly correct them within a viewer reduces discrepancies between the site and the data and contributes to greater efficiency in downstream workflows.

What we can do 9. Efficient operation through data reduction and optimization

While high-density point cloud data contains a large amount of information, its file sizes tend to become huge, which can make handling them difficult. For that reason, viewers and related tools often include features for reducing and optimizing point cloud data. The basic strategy is to use compressed file formats. The LAS format is a standard binary format, but if you use the losslessly compressed LAZ format, data size can be reduced to about one-tenth of the original. For example, a 500 MB LAS file can be compressed to about 50 MB as LAZ, making distribution by email attachment or via the cloud realistic. Moreover, because LAZ is a lossless compression method that yields data identical to the original LAS at the bit level when decompressed, it has the advantage of reducing size without any information degradation.

Next, data partitioning and filtering are effective. Instead of packing the entire site into a single file, dividing files by area or content lowers processing load because you can load only the parts you need for the task. Removing unnecessary points from the point cloud is also effective. Point clouds often include noise such as buildings, trees, vehicles, and people in addition to terrain, and removing these depending on the use case can reduce data volume. By using a viewer’s filter function to hide specific classes (e.g., non-ground surfaces) or trimming unrelated parts with a spatial clipping function, you can narrow down to only the information you want and display and analyze it responsively. Furthermore, cloud-based viewers employ server-side spatial partitioning and streaming technology that sequentially loads only the required parts with LOD (level of detail) control, enabling smooth display of point clouds on the order of hundreds of millions of points even on typical PCs and network environments. In this way, through data lightweighting and optimization techniques, heavy point cloud data can be made manageable and useful in everyday work.

What You Can Do 10. Smooth communication through sharing and collaboration on point cloud data

The tenth important feature of a general-purpose point cloud viewer is a mechanism that supports data sharing and collaboration. Traditionally, point cloud data acquired on site had to be taken back to the office for processing and conversion, and it could take days before being distributed to stakeholders. However, with a cloud-connected viewer, you can upload immediately after scanning on site and share instantly. For example, if you perform a laser scan at an excavation site and upload the point cloud data to the cloud, colleagues in the office can view the current 3D in their browsers within just a few minutes. Because the latest point cloud model can be shared even when people are remote, on-site personnel can request additional instructions or advice on the spot. In fact, there are cases where buried objects were found, the point cloud was shared immediately, and the office promptly decided whether additional excavation was necessary.

In this way, real-time information sharing closes the gap between the field and the office and dramatically accelerates decision-making. Because point cloud data in the cloud is centrally managed and always kept up to date, there is no worry that team members will refer to different, outdated files and cause discrepancies. By setting access permissions, you can share data with external clients and partner companies to the necessary extent while maintaining control over information. Some cloud services even provide real-time collaboration features that allow multiple users to access the same point cloud online simultaneously and annotate it together, as well as history and version control mechanisms. These sharing and collaboration features expand the use of point cloud data from mere records to communication tools that connect the field and the office.

Summary

So far we've looked at the various functions of point cloud viewers, but to make the most of them, the accuracy and timeliness of the source data are also important. For example, to reduce the effort of coordinate transformation and registration, it is ideal to acquire point clouds with high-precision absolute coordinates based on on-site control points from the outset. A recent solution that has emerged is "LRTK," an iPhone-mounted GNSS high-precision positioning device that can be used in combination with a smartphone. By combining this LRTK with a smartphone's LiDAR scan, it becomes possible to acquire point cloud data with centimeter-level positional accuracy (cm level accuracy, half-inch accuracy). Accuracy that previously required expensive GNSS survey instruments and laser scanners can now be achieved relatively inexpensively and easily with just a smartphone. If you can record high positional accuracy 3D data, it helps with as-built management (post-construction shape verification) and resolves uncertainties that photos tend to make ambiguous, such as "which point was measured" and "from what angle it was viewed." Also, because point clouds obtained with LRTK are aligned from the start to absolute coordinates such as public coordinate systems, the effort involved in post-processing coordinate correction and merging is greatly reduced. In short, by utilizing LRTK, anyone on-site can easily obtain high-precision, immediately usable 3D point clouds, allowing various functions of point cloud viewers to be more directly applied to practical work.