5 Ways to Improve the Accuracy of Absolute Coordinate Point Clouds — Explaining Causes of Errors

As the use of point cloud data in business increases, there is growing demand not only to view them in 3D but also for them to properly align with existing drawings, maps, design data, and point clouds acquired on different days. For practitioners in particular, a point cloud that is correctly placed in absolute coordinates and can be confidently overlaid with other data is far more valuable than one that merely looks good. In practice, however, there are many common problems: a point cloud that was supposed to be aligned to absolute coordinates is slightly offset from drawings, data from different days do not match, only the elevations appear different, or errors grow larger toward the edges of the site.

The reason such problems occur is that many factors are intertwined—not only the acquisition accuracy of the point cloud itself, but also how control points are established, understanding of the coordinate system, the selection of corresponding points, the local site environment, and the methods used to check post-processing. In other words, the accuracy of an absolute-coordinate point cloud is not determined solely by the performance of the equipment. Rather, the usability of the deliverables depends greatly on how the reference is secured on site, what procedures are used to achieve consistency, and how thoroughly verifications are performed.

When aiming to improve the accuracy of point clouds in absolute coordinates, many people first focus on software settings and transformation methods. However, merely revising processing without properly understanding the causes of errors will not lead to fundamental improvement. What is important is to identify where errors are likely to arise and to address their causes in sequence. By doing so, point clouds become not just three-dimensional data but practical data that can be readily used as survey deliverables and management documentation.

In this article, we first organize the basic concepts you should grasp to improve the accuracy of point clouds in absolute coordinates, and then explain five specific improvement methods. We also clearly describe the main causes of errors in absolute-coordinate point clouds. If you are planning to start using point clouds in your work, are concerned about the accuracy of point clouds aligned to absolute coordinates, or want to reduce re-measurements and backtracking in the field, be sure to read through to the end.

Table of Contents

• What is an absolute coordinate point cloud?

• Why accuracy matters for absolute coordinate point clouds

• Main causes of errors in absolute coordinate point clouds

• Method 1: Review the placement and number of control points

• Method 2: Improve field observation conditions and data acquisition procedures

• Method 3: Strictly select corresponding points on the point cloud

• Method 4: Enforce consistency in coordinate systems and height references

• Method 5: Refine final accuracy using check points and residual verification

• Practices beginners should avoid when trying to improve accuracy

• Summary

• What is an absolute coordinate point cloud?

An absolute-coordinate point cloud refers to a point cloud whose data are referenced to a coordinate system that can be used in common with other drawings, maps, and survey results, rather than to provisional site-specific positions or local coordinates. Immediately after acquisition, point cloud data are often stored in the measurement instrument’s proprietary internal coordinates or in relative coordinates within processing software, and even in that state it is possible to confirm the shape of the object and distance relationships. However, in that state it is difficult to accurately overlay them with existing plan drawings or point clouds acquired on different days, which limits their practical range of use.

For example, even if you generate point clouds of a building’s exterior perimeter or the terrain at a site, if that data remains in local coordinates the positions will not line up when you compare it with point clouds taken on another day or when you match it against existing registry drawings. Even if distances are correct within the point cloud itself, it cannot be placed in a common spatial frame with external data. In contrast, point clouds referenced to absolute coordinates can be handled with the same positional reference as maps and drawings, enabling smooth integration with other data.

What matters here is that absolute-coordinate point clouds are not merely an aesthetic issue. Being rendered nicely in 3D is quite distinct from having practical positional accuracy. No matter how carefully the field measurements are taken, if the coordinate system is confused or control points are shifted, the deliverable will be a point cloud that is difficult to use. Conversely, if the coordinate system is well understood and aligned to absolute coordinates, the point cloud holds great value as surveying results, management ledgers, maintenance and management materials, and comparative verification materials.

Also, when we talk about absolute coordinates, it is not enough that the coordinates simply have large numerical values. What matters is which coordinate system they are based on, what the reference for horizontal position and elevation is, and whether the assumptions needed to align with other data are in place. Confusion such as the horizontal position being correct but only the elevation differing, intending to handle data in latitude/longitude but actually operating in projected coordinates, or mistakenly believing a site-specific coordinate system is an absolute coordinate can occur not only among novices but also among experienced practitioners.

In other words, an absolute-coordinate point cloud is a state in which the prerequisites for connecting point cloud data to the outside world are in place. It is precisely because of this prerequisite that a point cloud transforms from a standalone 3D model into highly reliable data usable in business operations. If you are considering improving accuracy, it is important to first correctly understand what this means.

Why Accuracy Matters in Absolute Coordinate Point Clouds

The reason accuracy is important for point clouds in absolute coordinates is that point clouds are not data that are complete on their own; their value increases only when combined with other information. While keeping them in local coordinates allows for checking an object’s shape and obtaining rough measurements, most practical work does not end there. Point clouds are used on the assumption that their positional relationships to other data are correct—for tasks such as comparison with design drawings, verification of as-built conditions, comparison of changes over time, reconciliation with existing drawings, and integration of measurement results from multiple days. Therefore, if the accuracy of point clouds in absolute coordinates is poor, all subsequent processes will be affected.

For example, when comparing before-and-after construction, if the two point clouds are not aligned to the same coordinate reference with high precision, it becomes difficult to determine whether what appears to be a change is truly a change on site or merely a coordinate shift. This is a major problem in contexts such as maintenance management and displacement monitoring. Likewise, in the recording of cultural properties and archaeological remains, if the results from each survey area are not organized with common positional information, the materials cannot be correctly interpreted when reviewed later.

Furthermore, absolute-coordinate point clouds with insufficient accuracy cause mismatches in understanding among personnel. Even if the person who handled them on site has a rough grasp of the positional relationships, another person may feel something is off when overlaying them on drawings or maps. This then takes extra time to verify which is correct, and in the worst case may require remeasurement. Such rework often results in greater losses than the on-site work itself.

Accuracy also relates to the reliability of the deliverables themselves. Even if a point cloud looks neat, if the basis for alignment is unclear and residual checks are insufficient, it will be weak as documentation. Conversely, if it is clear which control points were used, which coordinate system was employed, and to what extent consistency was verified, that point cloud will be easier to explain to third parties. For those receiving the point cloud, it also becomes easier to judge how much they can trust and use it.

Another important aspect is that the higher the accuracy, the easier it is to reuse. In the field, there are occasions when a point cloud created once is reused for other projects, year-to-year comparisons, or checks during design changes. At those times, if the absolute coordinate accuracy is solid, it is easier to align with newly acquired data and the value of past deliverables is preserved for longer. In other words, improving accuracy not only raises the quality at that moment but also ensures future usability.

Main causes of errors in absolute-coordinate point clouds

To improve the accuracy of absolute-coordinate point clouds, you first need to understand where errors originate. Errors can be caused by a single factor, but in practice they are usually the result of several small factors accumulating into a larger one. If you don't sort this out and simply respond by "changing the settings because it seems a little off," it will not lead to a fundamental solution.

The first issue to mention is the error of the control points themselves. Since an absolute coordinate point cloud is placed into a coordinate system using known points or control points, if that reference is unstable the entire outcome will be unstable. Problems such as poor observation conditions for the control points, positioning results that are not sufficiently stable, markers being moved on site, or confusion in numbering and records will carry through after the transformation. Beginners tend to focus only on the conversion results from processing software, but if the underlying reference quality is poor, it is difficult to make up for that in later work.

The next most common issue is errors caused by how corresponding points are picked on the point cloud side. In the field, coordinates are assigned to the center of the target, but on the point cloud you might pick a point near the edge instead of the center, or intend to use a corner of a structure but select a different intersection. Even if these look like small misalignments visually, they affect the overall rotation and translation in the coordinate transformation, causing errors that increase toward the edges.

Confusion between coordinate systems and vertical datums is also a major cause. In point cloud processing, problems such as the horizontal coordinate system being correct while the height was handled using a different datum causing a Z-direction shift, mistakenly processing a coordinate system number as the projection for a different region, or treating latitude/longitude and projected coordinates ambiguously occur very frequently. Moreover, these issues are difficult to notice immediately after processing and often only become apparent as inconsistencies when overlaid with drawings or other deliverables.

Errors caused by on-site environmental conditions cannot be ignored. Obstructions from buildings and trees, limited sky visibility, reflections and multipath, swaying of targets due to wind or pedestrian traffic, and line-of-sight restrictions caused by scaffolding or equipment all affect both control point observations and point cloud acquisition. Any one of these may be minor on its own, but when several occur together accuracy can drop significantly. In particular, because ideal observation conditions are not always achievable in outdoor environments, measures appropriate to the situation are necessary.

Furthermore, there are error factors in the acquisition conditions of the point cloud itself. Biases in viewpoint, occluded areas of the object, insufficient point density, excessive noise, and monotonous surfaces with few feature points can destabilize alignment and the selection of corresponding points during post-processing. Before placing the point cloud into absolute coordinates, you must first confirm that it clearly includes the reference points and that the entire object has been stably captured.

Finally, insufficient verification after transformation is also a cause of overlooking errors. If you feel reassured by looking only at the mean residual, judge consistency only by the points used in the adjustment, or look only at planar positions without checking the vertical direction, problems will surface in later processes. Errors are not necessarily found where they occurred, and they are often only visible in the final inspection step.

Thus, the causes of errors in absolute-coordinate point clouds span multiple stages — control points, the field environment, point cloud acquisition, selection of corresponding points, understanding of the coordinate system, and insufficient verification. Therefore, to improve accuracy it is important not only to adjust processing software settings but also to review the preceding stages in order.

Method 1: Reevaluate the placement and number of reference points

One of the most basic and effective measures for improving the accuracy of an absolute-coordinate point cloud is to reassess the placement and number of reference points. Reference points are the foundation for placing the point cloud into an absolute coordinate system. If this foundation is weak, no matter how carefully you perform the transformation, the results will not be stable. Conversely, simply improving the approach to reference points can greatly enhance the accuracy of the entire point cloud.

First, it's important not to concentrate reference points only in locations that are convenient to work in. For example, if you secure reference points only on one side of the site or in the center and perform a coordinate transformation, the fit may look good in that area but there can be large discrepancies on the opposite side or around the perimeter. Because coordinate transformation of a point cloud affects the entire space, it is more stable to place reference points so they surround the target area as much as possible, balanced both planimetrically and three-dimensionally.

Next, it is important to ensure that the number of control points is not kept to an absolute minimum but secured with some margin. In theory, conversion may be possible with only a few points, but in practice it is safer to secure multiple points to allow for checks and the exclusion of outliers. However, more points are not necessarily better. Increasing the number of ambiguous points can actually disrupt the overall consistency. What matters is choosing points that can be reliably identified, whose coordinate values are highly trustworthy, and whose distribution is not biased.

The shape of a reference point also affects accuracy. Targets whose centers or intersections are easy to identify on the point cloud have higher reproducibility when selecting corresponding points. Conversely, objects that are very rounded, have indistinct corners, or have noisy surroundings tend to yield picked positions that vary between people. Considering, from the stage of installing reference points on site, whether they will be clearly readable on the point cloud later will improve the stability of the transformation.

Stability is also a point that is easy to overlook. On site, targets or markers may appear to be simply placed, but in reality they can move slightly due to human contact, wind, or ground conditions. Even a small movement can become a major problem as a reference for absolute coordinates. Especially for work that spans long periods or for sites involving multiple observations, it is essential to be vigilant in confirming that reference points have not moved.

Furthermore, it can be effective not to use all reference points for adjustment, but to keep some aside for validation. Even if the points used in the transformation calculations appear to match well, checking other points can reveal discrepancies. By separating points for adjustment and for validation, you can evaluate the transformation results more objectively. This may feel a bit difficult for beginners, but it is a very important concept for achieving stable accuracy.

A reference point is not merely a point with coordinates; it is the starting point that underpins the reliability of the entire point cloud. When the accuracy of the point cloud in absolute coordinates falls short of expectations, you should first reassess the number, locations, shapes, stability, and operational procedures of the reference points. Improving this alone will make subsequent processes much easier to carry out.

Method 2: Prepare on-site observation conditions and data acquisition procedures

Even if reference points are positioned appropriately, if on-site observation conditions or acquisition procedures are inconsistent, the accuracy of absolute-coordinate point clouds will not be stable. Therefore, the second important approach is to standardize the on-site observation environment and make the point-cloud and coordinate acquisition procedures as stable as possible. When it comes to improving accuracy, there is a tendency to rely on post-processing, but in practice differences in field conditions have a major impact on the final result.

First, be careful not to treat control point surveying and point cloud acquisition as entirely separate. It is not uncommon for coordinates to be measured accurately while the control points are not clearly captured in the point cloud, or conversely for the points to be well visible in the point cloud while the coordinate observations were unstable. To improve the accuracy of an absolute-coordinate point cloud, both must function together as an integrated whole. In other words, you should aim for the points used to assign coordinates to be clear and stable on site and included in the point cloud in a way that makes them easy to identify.

The next important factors are occlusion and visibility. If reference points are located in the shadow of a building, under trees, in the shadow of equipment, or behind temporary structures, this will be disadvantageous for both coordinate acquisition and point cloud capture. Even if you can place reference points, it may become difficult to determine the center from the point cloud, or they may be invisible when integrating multiple viewpoints. On site, it is important to consider not just whether you can place them, but whether they can be stably verified from multiple directions.

Also, the order of point cloud acquisition and the planning of viewpoints affect accuracy. If the viewpoint arrangement does not allow the entire subject to be covered without difficulty, some areas may have low point density or the point cloud around reference points may be unclear. Especially on large sites or with complex structures, rather than acquiring data haphazardly, thinking in advance about where to shoot from and in what order will make the results more consistent.

Furthermore, attention should be paid to changes in site conditions. Even at the same site, surrounding congestion, lighting conditions, equipment layout, and the visibility of the sky can vary depending on the time of day. Because these factors affect the stability of reference point observations and the visibility of point cloud acquisition, working at times when conditions are as stable as possible makes it easier to ensure accuracy. Although site constraints sometimes prevent choosing an ideal time, it is more important to plan acquisition procedures based on an understanding of how conditions will affect the results than to force progress under poor conditions.

Maintaining field notes should not be overlooked. If it is unclear which point corresponds to which number, which acquired data corresponds to which location, or at what timing observations were made, the correspondence can break down during post-processing. To improve accuracy, not only numerical precision but also precision in information management is necessary. If on-site records are well organized, it becomes easier to trace the causes of outliers and to perform readjustments.

In other words, organizing on-site conditions and acquisition procedures is a preprocessing step for improving accuracy. Even with advanced processing later, there are limits if the information necessary at the site is missing. If you want to increase the accuracy of absolute coordinate point clouds, it is essential to be conscious of reducing sources of instability during the on-site phase.

Method 3: Make the selection of corresponding points on the point cloud side more rigorous

The third method is to precisely select, on the point cloud side, the positions that correspond to known points. This step is particularly susceptible to human judgment during post-processing and can produce substantial differences in accuracy. No matter how well reference points are prepared on site, if the corresponding points are picked ambiguously on the point cloud, the final accuracy of the absolute-coordinate point cloud will deteriorate.

The first thing to keep in mind is that the location assigned coordinates in the field and the location you pick on the point cloud must always be the same physical point. This may seem obvious, but in practice it often shifts. For example, you might assign the coordinates to the center of a sign but use part of its outline as the reference on the point cloud, or intend to use the corner of a structure but actually select a different point along the face’s crease instead of the corner. Even if these discrepancies appear small, when you perform a transformation using multiple points they affect the overall rotation and translation and lead to a reduction in accuracy across the entire site.

Therefore, when selecting corresponding points, it is important to standardize rules for which part of a shape will serve as the representative point. For a circular target, use the center; for acute structures, use the intersection; for known markers, use the specified reference position—essentially, have criteria that ensure anyone will choose the same location. If different operators pick points differently, reproducibility is lost, which becomes a source of error in multiple projects and ongoing measurements.

Also, it's important to make judgments while checking the point cloud density and the noise level. If there are few points around a reference point, the contour is blurred by reflections or noise, or only part of it is visible due to a blind spot, that point is riskier than it appears. Beginners tend to think, "It's visible, so it can be used," but if you want to improve accuracy, excluding even slightly ambiguous points can sometimes yield better results. This process should prioritize quality over quantity.

Furthermore, attention must be paid to the distribution of corresponding points. Even if you select only high-precision points, if they are concentrated in a narrow area, errors tend to appear farther away. In other words, rigorous selection of corresponding points is not merely about carefully picking individual points; it also involves choosing them so they are meaningfully distributed across the entire space. It is important to adopt the perspective of stabilizing the entire point cloud.

Selecting corresponding points on the point-cloud side may look like it can be done in just a few clicks in the software. However, in reality it is a critically important judgment process that determines the accuracy of the point cloud in absolute coordinates. If this step is handled carelessly, the value of control point surveying and careful field work will be halved. To improve accuracy, you should spend time on the task of selecting corresponding points.

Method 4 Standardize coordinate systems and vertical datums

The fourth method is to thoroughly standardize the coordinate system and vertical datum. This step is indispensable not so much for improving the accuracy of the absolute coordinate point cloud as for creating a situation in which errors can be recognized as errors. This is because when coordinate systems or vertical datums are mixed, you cannot determine whether the observed discrepancy is truly a positioning error or merely a difference in reference.

In practice, the person handling point clouds and the person managing existing drawings are often different, and assumptions about the coordinate system may not be shared. For example, the plane rectangular coordinate system’s zone number may differ, the existing drawings may be managed in site coordinates, the drawings may be in meters while the point cloud is processed in a different unit, or only elevation may use a separate reference datum — small differences in understanding can appear as large discrepancies. Moreover, because the point cloud’s shape and relative relationships are not affected in these cases, it can be hard to notice just by looking at the processing software.

Height reference, in particular, is easy to overlook. While planar positions may appear to match well, the heights can be uniformly offset, or there can be significant vertical discrepancies when compared with drawings—these issues may stem from differences in the height reference. Beginners tend to focus on XY alignment and postpone the Z direction, but in 3D point clouds matching heights is also critically important.

Consistency is also important when transferring data. Even if the coordinate values are the same, other personnel cannot handle them correctly unless the coordinate system is specified. When saving converted point clouds, make it clear in the file name or in the management information which coordinate system and height reference they are based on so that operations remain stable. This is less an improvement in accuracy itself than a management practice for maintaining accuracy.

Also, you should check that the units have not changed before and after the conversion. Confusion between millimeters (mm / in) and meters (m / ft), reversal of axis direction, and differences in the orientation of east, west, south, and north can surprisingly occur at sites performing absolute coordinate alignment. Because these kinds of problems are difficult to judge by looking at the numbers alone, it is important to overlay them on existing drawings or known structures and verify them visually as well.

If you want to improve the accuracy of point clouds in absolute coordinates, it is important not to treat errors as solely a positioning or observation problem. If the coordinate system and vertical datum are not unified, no matter how finely you reduce the residuals their significance diminishes. First, establishing a condition in which comparisons can be made using the same measuring standard is a prerequisite for improving accuracy.

Method 5: Refine the final accuracy by checking validation points and residuals

The fifth method is to refine the final accuracy by using validation points and residual checks. This is the final step that determines whether the absolute-coordinate point cloud can be used with confidence in practical work. Even if the point cloud appears to be registered to the coordinate system, you cannot be sure it is truly high‑precision unless you objectively verify its consistency.

First, what you should check is how much deviation remains at each control point used in the transformation. In general, transformation procedures produce residuals for each point, but the important thing here is not to judge by the average alone. Even if the average is small, there may be a single point that is largely off. In such a case, it could be due to a mistaken identification of the corresponding point, an observation anomaly at the control point, or that the geometry of that point itself was inappropriate. Don’t be misled by the appearance of the average; it is important to examine the deviation of each individual point.

Another effective approach is to verify using check points that were not used in the transformation. Points used for adjustment tend to fit the calculations better, so if you evaluate accuracy using only those points you can end up with an overly optimistic assessment. If you confirm consistency using separately secured known points or existing structures whose positions are well defined, you can obtain an evaluation that more closely reflects reality. This is very effective for increasing the reliability of absolute-coordinate point clouds.

Also, it is essential to examine the planar and vertical directions separately. If the displacement in the XY directions is small but only the Z direction is consistently offset, issues with the vertical reference or observation conditions may be suspected. Conversely, if heights are fine but the plane appears rotated, the cause may be a biased arrangement of reference points or problems in selecting matching points. Rather than summarizing errors into a single number, reading the trends in each direction is the quickest way to improvement.

Additionally, it is important to visually verify the entire point cloud by overlaying it onto existing drawings or known structures. Even small numerical residuals can produce unnatural offsets at wall edges, road curbs, corners of structures, and similar features. Especially at the edges of the surveyed area or where there are changes in elevation, distortions that are easy to overlook from calculations alone tend to appear, so visually checking multiple locations provides added assurance.

Ultimately, you must determine how much error can be tolerated for the intended business purpose. While higher accuracy may seem better, in practice the balance with the purpose is what matters. The required level varies depending on whether it is for design verification, for gaining a general understanding, or for long-term comparison. However, as a prerequisite for setting tolerance ranges appropriate to the purpose, the residuals and validation results must be clearly organized.

The final push toward improved accuracy lies not in flashy processing but in careful, thorough verification. Ensuring validation points and residual checks are not skipped, and making necessary readjustments based on an understanding of deviation trends, is the key to producing an absolute-coordinate point cloud you can rely on.

Mindsets Beginners Should Avoid to Improve Accuracy

So far we have looked at ways to improve accuracy, but finally I will整理しておきます — wait, must not leave Japanese. Replace with: "but finally, let's organize the ways of thinking that beginners tend to fall into." Use natural: "but finally, let's go over the ways of thinking beginners are prone to." Okay produce final.

Correct final:

So far we have looked at ways to improve accuracy, but finally, let's summarize the ways of thinking that beginners tend to fall into. In real-world situations, it is often not technical shortcomings but habitual thinking patterns that prevent accuracy from improving.

First, something to avoid is assuming that a point cloud is accurate just because it looks clean. When an object appears smooth in a 3D view and the colors are well balanced, it's easy to assume the accuracy is high as well. However, for point clouds in absolute coordinates what matters is not appearance but whether they are correctly registered to the known coordinate system. Even if the visual appearance is good, if it doesn't match the drawings, the elevation is different, or datasets from multiple days do not align, the practical accuracy is insufficient.

Next, it is also dangerous to overestimate that errors can be resolved through software settings. Of course processing parameters are important, but many errors are determined by on-site control points, observation conditions, and the selection of corresponding points. If you try to force a match by post-processing alone, some areas may look fine while distortions appear elsewhere. It is safer to regard accuracy improvement as the accumulation of prior steps rather than as software techniques.

Also, you should avoid the mindset that you can manage with very few control points. That a transformation can be completed with a minimum number of points is not the same as achieving stable accuracy across the entire area. Not only the number but also the arrangement of points is important, and skewed control points are likely to cause discrepancies at the edges or in areas with elevation differences. If you skimp on control points to save effort, you may face many times the verification costs in later processes.

Being reassured by looking only at the average residual is a common mistake among beginners. Because the numbers are consolidated into a single value, it’s easy to judge quality based only on that. In reality, however, you can have one point that is significantly offset, the plane may fit but the height be off, or the center may be good while the edges are poor. It is important to make a habit of examining residuals not only by their average but also by individual points and by directional trends.

Furthermore, it's a mindset you should avoid: starting processing before you understand the coordinate system. If you proceed with conversions while leaving the coordinate system identifier, vertical datum, units, and assumptions about the drawing's coordinates ambiguous, you'll later be troubled by unexplained offsets. To improve accuracy, you need to align the measuring standards first.

What beginners should first develop is the sense that high-precision absolute coordinate point clouds are not produced by chance, but are the result of carefully building up preparation, observation, selection, transformation, and verification. Once you adopt this mindset, your perspective on accuracy will change significantly and you will become better at isolating problems.

Summary

To improve the accuracy of absolute-coordinate point clouds, simply performing the transformation process well is not enough. First, understand where errors originate, then review the placement and number of control points, refine field observation conditions and acquisition procedures, strictly select corresponding points on the point cloud side, unify the coordinate system and vertical datum, and finally finish by using check points and residual verification. These five methods are not independent measures but are all interconnected. Improving only one of them will not yield stable accuracy if the others remain unclear.

Especially for practitioners searching for information on "absolute-coordinate point clouds", it's important not to leave accuracy to the equipment or software. High-precision absolute-coordinate point clouds are determined by the overall capability that encompasses site preparation, the approach to reference control, care during acquisition, decisions made in post-processing, and final verification. Even if they look clean, they are meaningless if they don't align with other data. Conversely, if the rationale for each step is well organized, the point cloud becomes a highly reliable deliverable that can be used for a long time.

Also, if you want to stabilize the accuracy of absolute-coordinate point clouds, it’s important to review not only the point-cloud processing steps but also the on-site position acquisition itself. If you want to efficiently establish on-site reference points and link positions between photos and point clouds, using a method such as LRTK, an iPhone-mounted GNSS high-precision positioning device, can make it easier to establish the groundwork for absolute coordinates. For those who want to consistently align on-site position acquisition with post-processing, consider not only how you use point-cloud software but also introducing high-precision positioning like LRTK, as this can more readily help balance accuracy and operational efficiency.