All on your smartphone! Photogrammetry × Cloud for instant 3D capture of the site

How convenient would it be if you could instantly convert a site’s situation into 3D data and share it? In the construction industry, 3D surveying and digital construction management are being promoted through initiatives such as the Ministry of Land, Infrastructure, Transport and Tourism–led i-Construction, and the use of 3D models on site is becoming increasingly important. Until now, converting a site to 3D required special equipment such as expensive laser scanners or drones, but there has been growing demand to achieve site 3D conversion in a more accessible way. However, conventional methods required dedicated surveying instruments and advanced skills, and data processing took time. Even when point cloud data were obtained, they could not be shared immediately between the site and the office, causing time lags in consensus building and other issues.

What has attracted attention, therefore, is photogrammetry using smartphones and cloud services. Photogrammetry is a technique that reconstructs the three-dimensional shape of an object from multiple photographic images. In recent years, it has become possible to quickly convert a site into a 3D point cloud using photos taken with the high-performance cameras of smartphones. Because this method can be completed with just a smartphone and no special surveying equipment, site technicians and municipal staff can acquire 3D data as part of their routine tasks. This article explains the mechanism and advantages of photogrammetry using smartphone photography plus cloud processing, along with the benefits of real-time sharing, remote verification, and immediate feedback. It also introduces concrete use cases for surveying, construction, and inspection, and at the end of the article touches on a new simple surveying method using smartphones called LRTK.

Photogrammetry with a smartphone? How it works and how easy it is

Photogrammetry has long been used in aerial photogrammetry and drone photogrammetry, but it is now easily achievable with a smartphone. The principle is simple: a computer analyzes multiple photos taken of the object or site from various angles, identifies common features between the photos to estimate camera positions and shapes, and reconstructs a 3D model or point cloud of the subject. Previously, high-resolution cameras and dedicated software were required, but improvements in smartphone camera performance and computational techniques mean that the smartphone in everyone’s pocket can now serve as a 3D scanner.

When performing photogrammetry with a smartphone, you use a dedicated mobile app or cloud service. For example, some systems automatically extract frames from a video shot with the smartphone camera and upload them to the cloud, while others require taking several dozen still photos around the subject and sending them together. The important point is that expensive equipment and specialized skills are not necessary. Even non-expert surveyors can simply follow on-screen guidance and shoot, and one person can cover a large area in a short time. Cases have emerged where detailed point clouds are obtained from photo sets taken with a smartphone for everything from small structures to entire buildings and even parts of terrain. With the latest smartphones offering high camera resolution, you can also add color information (texture) to the point cloud and recreate a realistic 3D model that reflects the site as it is.

In recent years, some smartphones (particularly iPhone Pro models) have been equipped with LiDAR sensors, enabling real-time 3D laser scanning. LiDAR measurement is strong in immediacy because it can generate a point cloud on the spot, but its effective range is limited to a few meters (a few ft) and it lacks some detail in fine features. On the other hand, photogrammetry can be used on devices without LiDAR, and can capture complex shapes in high-density detail. When capturing a wide area at once, you need to plan the number of photos and shooting procedure, but by using LiDAR and photogrammetry as appropriate for the situation, the potential of 3D data obtainable from a single smartphone expands dramatically.

Fast 3D conversion with cloud processing: from shooting to point cloud data in minutes

Generating point clouds and 3D models from photos taken on a smartphone ordinarily requires significant computation. However, by leveraging cloud services, you can immediately send site-captured data to analysis servers and perform fast 3D conversion. Heavy processing that is difficult on a smartphone is handled on the cloud side using GPUs for parallel computation, dramatically reducing processing time. In practice, even from several dozen high-quality photos, high-resolution point cloud generation can be completed in just a few minutes to a dozen or so minutes. This allows data processing that used to be done on a PC over a long period to be completed within the site’s break time.

A typical workflow for smartphone photogrammetry is as follows:

• Capture (data acquisition) – Use a smartphone to photograph the object or site you want to convert to 3D from various angles. Common approaches are walking around while recording video or taking still photos with sufficient coverage. The key is to capture the subject with overlap from every direction so that the photos have enough common points.

• Upload – Use a dedicated app to send the captured data to a cloud server. You can upload directly from the site via mobile data or Wi‑Fi, eliminating the need to transfer to a PC.

• Automatic analysis in the cloud – Photogrammetric image analysis runs automatically on the server. The system extracts feature points from the many photos, aligns them, and generates 3D point clouds or mesh models. No manual operation is needed; AI and algorithms perform the shape reconstruction in the background.

• Point cloud data generation – After waiting a few minutes, the 3D point cloud data is complete on the cloud. The model is scaled using photo Exif information and location tags, and can be converted to a real-world coordinate system if necessary. The completed point cloud notifies the smartphone app, and you can preview it immediately.

• On-site immediate verification – The completed point cloud can be checked on a smartphone or tablet right on site. You can rotate and zoom the model immediately after generation to check details, assess initial quality, and identify any missing areas. If needed, you can take additional photos immediately to supplement the data.

Because shooting to point cloud conversion is entirely completed on site, there is no time loss from having to take data back to the office for processing. The acceleration of analysis through cloud processing is especially significant, and the speed of 3D data generation has dramatically improved. Once the site person starts shooting with their smartphone, the cloud returns results while they attend to other tasks, minimizing wait time. The short analysis time is powerful in situations requiring immediacy, such as disaster response or night work.

Strengths of cloud point clouds: real-time sharing and verification

A major advantage of 3D point cloud data generated in the cloud is that it can be managed and shared online as-is. Traditionally, using point cloud data required opening it on a PC with installed dedicated software, and transferring it from the site to the office was cumbersome. However, if you upload point clouds to a cloud platform, anyone can view the same data immediately via the Internet. For example, right after measuring on site with a smartphone, you can share the point cloud via the cloud with headquarters or the client and have them remotely confirm the site situation in 3D in real time.

Concrete benefits of cloud point cloud sharing include:

• Immediate consensus building: Since everyone can simultaneously view the latest 3D information captured on site, there is no time lag of “wait until data processing is finished…”. Supervisors and designers in remote locations can grasp the current situation almost in real time and quickly provide instructions or feedback. Seeing is believing: information that is hard to convey in drawings or text becomes obvious in a point cloud model. This significantly speeds up decision-making and reduces rework on site.

• Reduced additional measurements: Cloud-hosted point clouds are accessible from PCs and tablets via a browser, so if someone in the office notices after the fact, “We should have measured here too,” they can immediately request the site staff to take additional photos. In the past, people often had to return to the office and then go back out when they found something missing, but thanks to instant feedback, cases of postponing additional surveying to another day are reduced.

• No dedicated software required for sharing: Some cloud services automatically issue a shareable URL for a 3D viewer for generated point clouds. Using this, clients and staff in other departments without licenses can view and manipulate the 3D data in a browser without installing software. Anyone can intuitively check the site situation 360°, making it easy to distribute as explanatory material or to share during remote meetings while discussing the point cloud together.

• On-the-spot measurement and analysis: Point cloud data uploaded to the cloud is not just for sharing; it can be used immediately. For example, you can measure the distance between two arbitrary points in the point cloud, create cross sections, or mesh it to calculate volumes on site. On a smartphone or tablet screen, you can instantly check quantities such as “how many cubic meters is the fill?” or “how many millimeters (in) is the displacement?”, and immediately begin corrective work if necessary. Compared with the traditional flow of taking data back and having an analysis department prepare calculation sheets, the ability to produce numbers on site immediately is a major efficiency improvement.

As described above, using cloud-hosted 3D point clouds obtained by smartphone dramatically shortens the cycle of “measurement → data conversion → sharing → decision.” The site and office become seamlessly connected by digital data, enabling the sharing of the site at that moment regardless of location—a smart workflow for the DX era.

Use cases for smartphone 3D: instant digitalization of sites in surveying, construction, and inspection

Now let’s look at representative use cases by sector—surveying, construction, and infrastructure inspection—to see how smartphone + photogrammetry 3D point cloud generation proves useful in real-world scenarios.

Use in surveying operations: rapid site assessment and terrain data acquisition

Smartphone 3D scanning is proving useful for wide-area site surveys and terrain assessment. Even in cases that used to require multiple people and days to set up total stations and survey, a smartphone can allow one person to complete the work quickly. For example, there are cases where finished-shape surveys for development sites using tablet-based point cloud measurements were completed in just 30 minutes, greatly reducing working time compared to conventional methods. A 3D scan can acquire millions of points at once and record surface irregularities exhaustively, reducing rework due to “missed measurements.” Speed is essential, especially at disaster sites or for urgent surveys. If you can obtain a detailed 3D model of the site simply by shooting video around it with a smartphone, it may be possible to finish everything from emergency on-site inspection to drafting the drawings on the same day.

In addition, high-density point clouds obtained can be post-processed to generate contour maps or extract arbitrary cross-sectional shapes. Smartphone photogrammetry is being pilot-introduced for tasks such as mountain road surveying and river cross-section surveys, and by creating necessary survey drawings from quickly acquired point clouds, it contributes to shorter survey periods and cost reduction. As a means of labor-saving surveying, photogrammetry using smartphones is likely to become even more widespread.

Use in construction sites: as-built management and progress sharing

Smartphone 3D conversion is useful in many aspects of the construction phase. A representative application is for as-built management. When confirming whether completed structures or ground shapes match the design, conventional practice was to manually measure key dimensions and compare them with drawings, but point clouds allow surface-level verification of the entire object. For example, in road embankment work, you can acquire the entire embankment slope using an iPhone’s LiDAR scan plus photogrammetry, and by overlaying the obtained 3D point cloud with the design 3D model, excesses and shortages are color-coded and immediately identifiable. Based on that information, you can add missing soil or cut excess areas on the same day, bringing the as-built into the allowable range before inspection. Performing pre-checks with point cloud heatmaps from smartphone scans prevents rework later and contributes to improved quality.

Also, leveraging cloud sharing to immediately share site progress is another construction management benefit. By photographing the site with a smartphone at each milestone and converting it to a point cloud, you can report progress in real time to remote office managers and clients. Progress sharing that was once done by emailing photos can now be done with 3D point clouds that convey the site conditions with a sense of presence, reducing communication loss. For example, you can 3D-record reinforcement placement in bridge construction and later compare it with drawings to check for construction errors. In this way, the easy on-site 3D scanning with smartphones is becoming a powerful tool to drive DX (digital transformation) in construction management.

Use in infrastructure inspection: detailed remote observation and record accumulation

Smartphone photogrammetry also offers new approaches in infrastructure inspection for roads, bridges, tunnels, and the like. If inspectors photograph the surfaces of structures with a smartphone, specialist engineers in the office can analyze the high-resolution 3D model, enabling efficient diagnosis and second opinions without the need for specialists to visit the site. For example, cracks in bridge piers or delamination of concrete surfaces can be measured on the point cloud or observed in detail using photographic textures. Also, deformations of tunnel lining concrete or rutting of roads—displacements that are hard to notice with the naked eye—can be quantified, aiding preventive maintenance.

Another major advantage is that acquired point cloud data can be retained as records for maintenance management. By storing infrastructure condition data that was previously managed with paper records or photo ledgers as 3D data, you can track changes over time. Questions such as “Has this crack widened compared to several years ago?” can be quantitatively evaluated by comparing past point cloud models with the present. Point clouds easily obtained with a smartphone and the cloud become a digital archive for infrastructure maintenance, contributing to long-term safety management.

Conclusion: Further efficiency with LRTK to accelerate smartphone surveying

Photogrammetry using smartphones and the cloud has made 3D measurement and point cloud generation, which used to be time-consuming and labor-intensive, easily accessible to anyone. This approach, which combines ease of use and speed, is a powerful ally in the digitization of construction and surveying operations. A workflow that enables immediate 3D capture on site and remote sharing and feedback of that data delivers efficiency that overturns conventional practices. Once actually used, many site technicians will likely feel, “I can’t go back to the old manual ways.”

Recently, attention has also focused on LRTK (smartphone RTK surveying) as a technology that expands the potential of smartphones. LRTK is a solution that attaches a small high-precision GNSS receiver to a smartphone and enables centimeter-level positioning using the RTK (Real Time Kinematic) method. Simply put, it’s like “turning a smartphone into a high-precision surveying instrument.”

By using LRTK, tasks that previously required dedicated equipment and tripods—such as control point surveying—can be performed with a smartphone immediately upon arriving at the site. You can obtain high-precision coordinates with errors of several centimeters (several in) simply by walking while holding the smartphone, then upload and share them to the cloud. Moreover, combining LRTK positioning data with 3D models generated by photogrammetry enables advanced uses such as attaching accurate position coordinates to point clouds for smooth comparison with design drawings.

By adopting technologies like smartphone photogrammetry and LRTK, site surveying and measurement will become increasingly simple and fast. If “surveying completed with a smartphone” becomes the norm, even limited personnel will be able to quickly and accurately capture sites, enabling efficient construction management and infrastructure maintenance. If your site has not yet introduced these methods, why not take this opportunity to try smartphone-based 3D measurement and simple surveying? With the latest technologies as your ally, you have a chance to dramatically improve on-site productivity and safety.