Table of contents

• What is a browser point-cloud viewer?

• Visualizing point-cloud errors with a heat map

• Benefits of no installation & cloud sharing

• Simple high-precision surveying with a smartphone: using LRTK

• FAQ

In recent years, usage of 3D point cloud data (point clouds) has been increasing, especially in civil engineering and construction. Entire sites are being scanned with laser scanners or drone photogrammetry, supporting as-built management and topographic surveying. However, there are challenges in sharing acquired point cloud data with people off-site and analyzing it in an easy-to-understand way. Dedicated software may be required, and large data volumes can be difficult to handle, so 3D information is often not shared fully. For example, point cloud data obtained from drone photogrammetry can reach tens of millions of points, and file sizes can be several GB. Such large files are hard to share by email and can be a struggle to open even on a typical PC.

What is attracting attention, then, is cloud-based data sharing and visualization using a browser point-cloud viewer. By uploading 3D point clouds to the cloud via the Internet and making them viewable in a web browser, 3D data can be exchanged easily with no installation required. Furthermore, by overlaying point cloud data with design data to create a heat map (a colored error distribution map), you can instantly judge whether the on-site as-built matches the design. This article explains the latest trends in browser-based point-cloud viewers and heat map usage. Finally, we introduce simple high-precision surveying using smartphones combined with a compact GNSS receiver, LRTK. If you are interested in digital transformation on site, please refer to this article.

What is a browser point-cloud viewer?

First, let’s define what a browser point-cloud viewer is. Traditionally, displaying and processing 3D point cloud data required dedicated desktop software or CAD programs. But recent advances in technologies such as WebGL allow 3D graphics to be rendered directly in a web browser. This has led to online point-cloud viewers where a browser loads point cloud data over the Internet and displays and manipulates it on-screen.

The biggest advantage of a browser point-cloud viewer is that no dedicated software installation is required. Viewers only need a web browser, so even those without special software or expensive workstations can view 3D point clouds simply by clicking a URL link. Whether on Windows or Mac, from an office PC or a tablet out in the field, the same data can be checked regardless of environment. Viewing, rotating, zooming, cross-section slicing, and distance measurements can all be done intuitively in the browser. Even without specialist knowledge, users can move the viewpoint with mouse or touch operations, making it easier for managers or clients who have never visited the site to imagine the 3D data.

Browser point-cloud viewers are also well suited to cloud-based data sharing. By uploading huge point cloud datasets captured on site to a cloud server and viewing them with a browser viewer, there is no need to exchange files each time. Since the latest data is always in the cloud, multiple team members can access and check it at their convenience. This enables sharing site conditions in real time across geographically dispersed locations and makes remote on-site attendance (remote inspection) possible for checks and meetings that previously required a physical presence. In fact, the Ministry of Land, Infrastructure, Transport and Tourism has been promoting institutionalization of remote inspections, and cloud sharing of point cloud data is expected to support this new inspection style.

Driven in part by initiatives such as i-Construction promoted by the Ministry of Land, Infrastructure, Transport and Tourism, open platforms and viewer technologies for handling point cloud data in the cloud are emerging one after another. Options range from open-source viewing libraries to commercial cloud services, allowing choices according to use case and budget.

For example, open point-cloud formats such as LAS/LAZ can scale to tens of millions of points and are optimized for smooth browser performance using spatial indexing and LOD (level of detail) techniques. Even without a dedicated in-house CAD operator, site staff can realistically scan and upload with a tablet and immediately share 3D data with office stakeholders.

Visualizing point-cloud errors with a heat map

Next, let’s look at heat map visualization of point cloud data. Simply looking at a point cloud by itself makes it difficult to intuitively grasp “where things match the design and where they deviate.” What is used here is analysis by a heat map comparing point clouds with design data. A heat map overlays measured post-construction point clouds with a design model (or a reference plane created from drawings) and expresses the vertical deviation at each point using color. For each point on the point cloud, the difference is calculated—e.g., “this point is protruding by ○ mm above the design surface” or “this point is ○ mm lower”—and displayed with colors such as red, yellow, green, and blue according to the magnitude of the error.

For example, in embankment works, areas where soil has been overfilled above the specified elevation or in cut areas where excavation is insufficient and the surface is raised above the design will appear red or orange on the heat map. Areas finished almost to the design will be shown in green to blue tones, allowing defective and satisfactory areas to be distinguished at a glance. The color gradient visually conveys error magnitude, so trends in construction accuracy—such as “overall slightly higher than specified” or “a specific area is lower than the rest”—are easy to understand.

Using heat maps allows detection of slight unevenness and small bumps that would previously have been overlooked. Traditional as-built inspections compared measured values at surveyed points against design values for pass/fail judgments, but heat maps enable area-based pass/fail checks across the entire space. Interpretation of results is also easy, and because color-coded 3D information is intuitive for everyone from site workers to clients, it becomes easier for the whole team to share where rework is needed.

More recently, AR (augmented reality) technology has appeared that overlays heat maps onto tablet or smartphone camera views. By downloading cloud heat map data to a mobile device and overlaying it in real time on the actual structure or ground at the site, you can confirm “which location needs how much correction” on the spot. Previously, marking defective areas found on a heat map required returning to the site and marking them manually, but with AR display the location of repair areas and the amount of deviation are visible through the screen, enabling immediate corrective work. By combining point-cloud heat maps with AR in this way, as-built evaluation is evolving from mere record-keeping into a real-time quality improvement tool.

Benefits of no installation & cloud sharing

As shown above, using a browser point-cloud viewer and heat map analysis dramatically improves the efficiency and sophistication of on-site 3D data utilization. Finally, let’s summarize the concrete benefits.

• Wide-area rapid measurement that shortens work time: Using 3D scanners or drones, even large sites can be scanned in a short time with high-density point clouds. A single scan can collect millions to tens of millions of points, covering areas that would previously have taken a full day to measure.

• Area-based measurement to prevent oversights: Point cloud data measures every surface detail of the object, so partial unevenness or distortion will not be missed. Tiny bumps that could not be noticed by traditional representative-point measurements can be detected by heat maps.

• Safe, non-contact measurement of hazardous areas: Laser scanners and photogrammetry can measure steep slopes, heights, or narrow spaces where people cannot safely enter, from a distance. This eliminates the need for hazardous work and improves on-site safety.

• Streamlined record-keeping and information sharing: As-built inspection results using point clouds can be digitally recorded in the cloud and shared immediately among stakeholders. The burden of creating paper photo books and extensive reports is reduced, and office and site can exchange information in real time. Because no special software is required and anyone can view 3D models in a browser, explanations to clients and internal sharing are smooth.

By incorporating 3D point clouds and heat maps into site management, inspections become more accurate, faster, safer, and less labor-intensive. Reducing rework due to measurement omissions and enabling all stakeholders to share the same visual information are major advantages over traditional methods. Point cloud data is not just a pass/fail record; it becomes a valuable 3D asset useful for future maintenance. If data is stored in the cloud after project completion, it can be used as reference material for future renovations or inspections.

Acquired 3D point clouds are also useful for electronic delivery of as-built results and for linking with BIM/CIM models. Delivering the completed structure’s point cloud model as data, rather than just paper drawings or photos, leaves a detailed record in the as-built documentation. Accumulated point cloud assets can be used for future maintenance and renovation planning and remain valuable information assets after construction is finished.

Point cloud visualization and sharing are also expanding beyond construction. Use cases where the convenience of sharing point clouds in a browser is valued include factory equipment maintenance and plant inspections, digital archiving of cultural properties, and accident scene documentation and analysis. An environment where anyone can view 3D models without specialized software is a powerful DX tool across industries.

Simple high-precision surveying with a smartphone: using LRTK

Finally, we introduce LRTK as a cutting-edge tool for easily acquiring point cloud data. LRTK is a compact high-precision GNSS receiver developed by a venture company originating from Tokyo Institute of Technology. By attaching it to a smartphone or tablet, centimeter-level positioning (cm level accuracy (half-inch accuracy)) that previously required specialized equipment becomes achievable for anyone. Combined with the LiDAR scanner function built into the latest iPhone or iPad, a smartphone can effectively become a high-precision 3D scanner.

In practice, simply walking around a site with an LRTK attached to a smartphone can record terrain and structure point clouds with good accuracy within an error range of several cm (several in) in some cases. With a dedicated app, the procedure is as simple as pressing a button; the measured positions and acquired point clouds are automatically uploaded to the LRTK cloud. Even in mountain areas or underground spaces without Internet coverage, LRTK can achieve high-precision positioning by utilizing satellite augmentation signals (for example, QZSS’s CLAS). Because the resulting data can be shared directly via the cloud, seamless workflows such as on-site measurement -> immediate cloud sharing -> remote heat map checking have become realistic.

LRTK has attracted attention on TV programs and at tech exhibitions, and adoption by local governments and construction companies is progressing. For example, in emergency surveying of terrain after heavy rain disasters, a smartphone with LRTK allows rapid assessment of current conditions without waiting for heavy machinery. It performs well in mountain areas or remote islands without Internet connectivity, enabling easy 3D measurement in scenes that were previously difficult.

Thus, the combination of a smartphone and LRTK is attracting attention as a simple one-device-per-person surveying solution. Without heavy tripods or special equipment, anyone can record a site in 3D and share it in the cloud. Even those new to point cloud usage can start with simple smartphone-based methods, greatly lowering the barrier to on-site DX. By using LRTK, the entire process from acquiring point cloud data to visualizing and sharing it in a browser viewer can be executed quickly and smartly.

FAQ

Q: Is it difficult to introduce a browser point-cloud viewer on my own?

A: No, it has become relatively easy to use in recent years. You can obtain open-source point-cloud viewer libraries and integrate them into your own servers, or use cloud services where you simply upload data to get a 3D viewer. With the latter, you can start using the service by registering an account without specialist knowledge, so it’s a good idea to try a small-scale free plan first.

Q: What data formats can browser point-cloud viewers handle?

A: Most viewers can read common output formats from laser scanners or photogrammetry. Browser viewers that support standard point-cloud files such as LAS, LAZ, PLY, and E57 are increasing. For proprietary formats, dedicated conversion tools or plugins may allow loading in some cases.

Q: Can large point-cloud datasets be displayed smoothly?

A: Recent point-cloud viewers are designed to visualize datasets in the tens of millions of points smoothly. Streaming technologies that spatially partition data and load only the needed parts, and LOD (level of detail) processing that reduces point detail based on display distance, allow smooth browser operation. WebGL enables fast rendering using the GPU built into PCs and tablets, so a typical laptop can handle large point clouds without stress. However, for extremely large datasets, pre-sampling (thinning) or tile partitioning will make viewing even more comfortable.

Q: Do I need design data to compare with a heat map?

A: To rigorously analyze as-built errors, you need reference data such as design drawings or a BIM model. Prepare 3D design data or a reference plane to compare against the point cloud and generate a heat map. If no reference data is available, you can still create a pseudo heat map by color-coding the point cloud itself by elevation, for example, to visualize terrain undulations. Choose the approach according to your purpose.

Q: Can smartphones be used for high-precision point-cloud surveying?

A: Modern smartphones with LiDAR sensors can easily obtain millions of points at short range. For phones without LiDAR, photogrammetry (SfM) techniques generate 3D point clouds from multiple photos. Using dedicated apps or cloud services, high-density point clouds can be produced from photos alone. However, standalone smartphones have limited positioning accuracy, so for professional surveying some measures are needed. This is where high-precision GNSS aids like LRTK are useful. Combined with a smartphone, they provide accurate coordinate references to acquired point clouds, making centimeter-level accuracy point-cloud surveying accessible.

Q: Is cloud-stored data secure?

A: Many cloud services ensure data security through encrypted communications and access control settings. Sensitive data that should not be publicly available can be protected with password-protected viewers or time-limited sharing links. When handling confidential point cloud data, choose a trusted platform and consider data encryption or contractual confidentiality as needed. Some also argue that centralized cloud management reduces the risk of loss or leakage compared with carrying data on physical media like USB flash drives.

Q: Can a browser viewer measure distance, area, and volume?

A: Many point-cloud viewers include basic measurement tools. You can measure the distance between two points on screen, slice sections from point clouds to calculate area, or estimate the volume of soil within a bounded region. However, formal quality inspections may require precision measurements using dedicated equipment, so browser-based measurements should be used as a guideline.