INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 637
Evaluating Point Cloud Measurement Accuracy for Residential
Property Valuation: A Case Study Using LiDAR Scanning
Rabi’atul’Adawiyah Azmil, Suzanna Noor Azmy, Siti Zaleha Daud
Faculty of Built Environment and Surveying, Universiti Teknologi Malaysia
DOI: https://doi.org/10.51583/IJLTEMAS.2025.1410000080
Received: 10 October 2025; Accepted: 20 October 2025; Published: 11 November 2025
Abstract—The most important part of property valuation is accurate property measurement. The more accurate measurement of
all types of property, including residential buildings, could lead to an equal and fair value for the property. By comparing linear
wall-to-wall measurements taken from point clouds with those taken from certified floor plans, this study evaluates the potential of
three-dimensional (3D) point cloud data for use in property measurement. Dimensional measurements were obtained by processing
3D data of a residence in Kajang, Malaysia, using LiDAR-based scanning via PolyCam Pro on an iPhone 14 Pro Max. To assess
the workflow’s generalizability, a sample dataset provided by PolyCam representing a landed residential unit was also tested using
the same measurement procedure. The results indicate that consumer-grade point cloud data can achieve accuracy sufficient for
expert valuation support, with deviations of approximately ±10 mm in the primary dataset and ±3.5 mm in the validation dataset.
This shows how point cloud technology can improve transparency, reduce measurement errors caused by individuals, and help in
the digital transformation of valuation workflows
Keywords: Point cloud, Measurement accuracy, Residential property valuation, LiDAR scanning, Floor plan comparison
I. Introduction
Residential property valuation depends on precise measurements of property features such as usable floor areas and wall dimensions [1].
However, traditional methods like manual tape-based inspection or reliance on architectural floor plans often introduce human error and
inconsistencies [2]. These measurement differences can accumulate throughout the valuation process, potentially affecting the final assessed
value and reducing professional credibility [3].
Recent developments in 3D sensing technologies particularly LiDAR and photogrammetry, enable the capture of highly detailed geometric
data in the form of point clouds [4], [5]. Point cloud data allow for objective, repeatable measurements by digitally representing the real-
world geometry of built environments. This capability improves measurement reliability and optimizes workflows in property valuation,
though its integration into valuation practice remains at an exploratory stage [6].
The objective of this study is to determine whether wall measurements based on 3D point clouds meet professional accuracy tolerances for
use in residential property valuation. The implications of using such data to enhance the accuracy and transparency of valuation measurement
procedures are also examined.
Background on Point Cloud Applications
Point cloud data are widely used in geoinformatics, engineering, and architecture for 3D reconstruction, infrastructure mapping,
and building documentation. With millions of spatial points captured, it offers a comprehensive geometric representation of built
environments [4], [5]. [7] confirmed the spatial accuracy potential of 3D city models using point clouds, while [8] demonstrated
that handheld LiDAR scanners can achieve centimeter-level accuracy suitable for indoor mapping and modeling.
Accessibility has improved through the introduction of mobile LiDAR, particularly in smartphones [9]. Professionals can now
collect geometric data efficiently using consumer devices instead of costly terrestrial laser scanners. Although sensor resolution is
lower than that of professional systems, studies have shown that consumer-grade devices can still provide reliable and consistent
results when processed appropriately [10].
Measurement Accuracy Studies
Measurement accuracy remains critical in both construction and valuation practices. Previous studies ([7], [8], [10]) have shown
that under controlled indoor conditions, low-cost LiDAR scanning technologies can maintain acceptable error margins, often within
±2%. Under Malaysian property and valuation standards, a ±2% difference in floor area measurements is permitted [12]. Therefore,
testing the precision of wall-to-wall measurements derived from point clouds is essential to determine whether this technology can
replace manual tape-based methods in valuation workflows.
Applications in Property Valuation
Over the past decade, the integration between valuation practice and geospatial technology has expanded considerably. [13] and
[14] explored how combining GIS and Building Information Modelling (BIM) enhances property data management and valuation
accuracy. [15] emphasized that 3D property information introduces volumetric data beyond traditional 2D boundaries, improving
the understanding of built form and value relationships.
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 638
However, the application of such advanced models in practice remains limited due to their reliance on automated extraction
workflows and complex data preparation. [17] proposed combining BIM and machine learning for automated valuation, yet this
requires significant computational expertise and data preprocessing. Direct measurement from point cloud data, by contrast, offers
a more accessible and cost-effective approach to improving measurement accuracy in property valuation.
II. Methodology
Study Area and Data Source
A residential apartment unit in Kajang, Selangor, Malaysia, was selected for this study because of its standardized internal layout,
accessibility for scanning, and availability of verified architectural floor plans. The PolyCam Pro mobile application was used on
an iPhone 14 Pro Max equipped with a built-in LiDAR sensor to generate dense three-dimensional point clouds in .ply format. This
consumer-grade device was chosen for its affordability, portability, and demonstrated ability to capture high-resolution spatial data
[9]. To maintain even coverage and minimize occlusion, the scanning was performed at approximately eye level with smooth
horizontal motion around the interior of the unit. Adequate indoor lighting was ensured during the capture process, although
variations in lighting and scanning angles were not quantitatively assessed in this study. Fig. 1 shows the overall scanning coverage
and geometric completeness of the captured unit.
Fig. 1 3D point cloud model of the residential unit captured using PolyCam Pro on an iPhone 14 Pro Max
Point Cloud Processing and Measurement
PolyCam Pro automatically merged multiple scans to produce a metrically referenced point cloud model of the entire internal geometry.
Wall intersection points were visually identified for use in subsequent linear distance measurements.
For measurement and visualization, the point cloud was exported from PolyCam in .ply format and imported into CloudCompare. Because
PolyCam automatically generates scaled data, no additional scaling, denoising, or filtering was required under the controlled indoor
conditions.
After import, the point cloud was visually inspected to confirm completeness and alignment. Minor edge noise was observed but found
negligible, and therefore no filtering or subsampling was applied. The coordinate scale generated by PolyCam was validated against a known
wall-to-wall distance measured on-site, confirming metric accuracy and minimizing potential systematic errors during measurement.
Using CloudCompare’s distance tool, wall-to-wall linear distances were extracted directly between defined wall intersection points. The
same process was repeated across multiple wall segments to ensure consistency and allow direct comparison with the architectural floor
plan.
Fig. 2. Wall-to-wall measurement extraction in CloudCompare using the distance tool
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 639
Fig. 2 illustrates the extraction of wall-to-wall measurements in CloudCompare using the distance tool and the complete point-cloud
processing and measurement workflow is summarized in Fig. 3.
Fig. 3. Point-cloud processing and measurement workflow
Accuracy Assessment
Measurements from the architectural floor plan served as the reference benchmark. The percentage difference (PD) between floor-
plan and point-cloud-derived measurements was calculated as:
PD =
| ������ − ������ |
������
�� 100
where ������ is the reference floor-plan measurement and �������� is the corresponding point-cloud measurement. PD was chosen
instead of Root Mean Square Error (RMSE) because the objective was to assess dimensional agreement rather than model-fit
accuracy.
Average deviations were approximately ±10 mm, corresponding to the reference wall lengths. These values fall within the tolerance
range (±2 %–5 %) recognized by Malaysian valuation and architectural measurement standards [12].
To determine whether mobile LiDAR scanning meets these professional thresholds, the deviations were evaluated against the
accepted measurement accuracy criteria. The analysis confirmed that consumer-grade LiDAR can provide measurement precision
adequate for professional property valuation tasks under controlled conditions.
III. Results
Primary Measurement Comparison
The dimensional differences between the architectural floor plan and the LiDAR-derived point cloud remained well within the
limits of professional tolerance. The average of roughly ±10 mm, is minimal and generally not significant in valuation practice,
approximately 5 cm over a 3.5 m wall.
These findings agree with the results of [8] and [11], who demonstrated that portable LiDAR systems can achieve sub-centimetre
precision in indoor settings.
Table I summarizes the comparison between floor-plan and LiDAR-derived measurements across the selected wall segments while
Fig. 4 illustrates the overall floor plan dimensions and measurement locations used in the comparison.
TABLE I. Comparison of wall-to-wall measurements between floor plan and point cloud
Wall ID Floor Plan
(Centerline) (m)
Floor Plan
(Offset
Adjusted) (m)
Point Cloud
Measurement (m)
Difference vs.
Centerline (m)
Difference vs.
Offset (m)
W1 (Master
Bedroom)
3.950 3.800 3.793 0.157
0.007
W2 (Master
Bedroom)
3.000 2.850 2.858 0.142 0.008
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 640
W3 (Bedroom
2)
3.000 2.850 2.854 0.146 0.004
W4 (Bedroom
2)
3.950 3.800 3.796 0.154 0.004
W5 (Bedroom
3)
2.700 2.550 2.562 0.138 0.012
W6 (Master
Bath)
2.400 2.250 2.237 0.163 0.013
W7 (Yard) 1.500 1.350 1.340 0.160 0.010
W8 (Kitchen) 6.100 5.950 5.944 0.156 0.006
Fig. 4 Floor plan dimensions
Model Validation Using Secondary Dataset
To evaluate the generalizability of the proposed workflow, a sample scan dataset provided by PolyCam was used for validation.
The dataset represents a landed residential unit, which differs from the primary apartment case in overall layout and scale. The same
point cloud processing and measurement workflow described in Section II was applied without modification.
Fig. 5 Measurement extraction in CloudCompare using the distance tool
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 641
Measurements were extracted in CloudCompare from the sample scan’s point cloud and compared with the automatically generated
floor plan dimensions supplied by PolyCam’s built-in floor plan generator.
Fig. 6 Floor plan dimensions generated by PolyCam
The validation results showed an average deviation of approximately ±3.5 mm, with the highest recorded difference being 11 mm.
These deviations are smaller than those observed in the primary dataset (±10 mm) and remain well within the ±2% tolerance defined
by professional valuation standards [12]. The minimal variation confirms that the workflow maintains consistent geometric
precision even when applied to a different residential property type. This finding demonstrates that the approach is robust,
repeatable, and generalizable across varied datasets, including those not directly captured by the researcher.
TABLE II. Validation results comparing PolyCam-generated floor plan and CloudCompare measurements from the sample dataset
Wall ID Floor Plan (m) Point Cloud Measurement
(m)
Difference (m)
W1 (Bedroom) 2.970 2.970 0.000
W2 (Bedroom) 3.390 3.391 0.001
W3 (Bedroom + Bathroom) 6.810 6.813 0.003
W4 (Bathroom) 1.480 1.478 0.002
W5 (Living room) 1.400 1.411 0.011
W6 (Living room) 4.310 4.312 0.002
W7 (Living room + Other 1) 8.310 8.312 0.002
W8 (Other 1) 4.000 4.003 0.003
IV. Discussion
The findings show that consumer-grade LiDAR sensors can produce 3D point cloud data with dimensional precision appropriate
for measurements related to valuation. Based on the Malaysian Valuation Standards (MVS) and the Royal Institution of Chartered
Surveyors (RICS), the observed average deviation of approximately 10 millimetres lies comfortably within the tolerance limits
generally accepted in professional valuation measurement [12].
A comparison between the primary dataset (apartment unit) and the validation dataset (landed residential unit) revealed consistent
measurement accuracy across both cases, with average deviations of ±10 mm and ±3.5 mm, respectively. The slightly smaller
deviation in the validation dataset indicates that the workflow remains stable and can even achieve improved precision when applied
to other residential property types. Both results remain well within the ±2% professional tolerance, confirming that the workflow
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 642
performs reliably and demonstrates strong generalizability across different datasets.
Implications for Valuation Practice
The application of point cloud data introduces measurable improvements in valuation accuracy and transparency. Unlike manual tape-based
methods, point cloud measurement offers objective, repeatable, and verifiable results independent of individual skill or bias. This improves
reliability and aligns with the emphasis on accountability and transparency in modern valuation governance frameworks [2].
Furthermore, 3D spatial data provide context often absent from 2D floor plans, enabling valuers to verify as-built conditions, assess spatial
relationships, and detect irregularities remotely. This capability supports emerging practices such as digital valuation models and remote
inspections, which have become increasingly common during and after the COVID-19 pandemic [14].
In practical terms, an average dimensional deviation of approximately 10 millimetres could translate to about 1.5% variance in reported
floor area, which in typical Malaysian residential markets corresponds to a value difference of roughly RM 3,000–5,000 for an RM 300,000
property [12]. Although this falls within acceptable valuation error margins, it highlights the importance of adopting standardized digital
measurement procedures in professional practice.
Limitations and Considerations
The study acknowledges several limitations The analysis was primarily based on one apartment unit, complemented by a validation test on
a landed residential dataset, which still limits the statistical generalizability of the findings to a broader range of housing types and
environments. Environmental factors such as lighting, surface reflectivity, and scanning angles were not quantified, although care was taken
to maintain stable conditions during scanning. Minor geometric distortions may arise from these influences or from device movement during
capture.
Additionally, the manual extraction of wall measurements from point clouds requires technical proficiency in software such as
CloudCompare, which may limit immediate adoption by valuers without geospatial training. Nonetheless, with standardized procedures,
point cloud data can provide a reliable and efficient alternative to manual measurement in residential valuation workflows.
Critical Reflection
This study contributes to the growing body of research on digital transformation in valuation, offering an empirical validation of consumer-
grade LiDAR accuracy for professional measurement. While prior studies have emphasized automated 3D reconstruction and BIM
integration [13], [17], the present work demonstrates that manual point cloud interpretation can still achieve measurement accuracy suitable
for professional standards under controlled conditions.
For future research, expanding the analysis to multiple case studies across different housing types and environmental conditions would
strengthen statistical validity. A sensitivity analysis examining how geometric deviations affect valuation outcomes (in percentage or
currency terms) would further enhance practical relevance. Future work may also explore semi-automated workflows integrating point
cloud processing with valuation software to reduce processing time while maintaining verifiable accuracy.
V. Conclusion
This study evaluated the measurement accuracy of consumer-grade LiDAR point cloud data for residential property valuation and
confirmed that dimensional deviations of approximately ±10 mm fall comfortably within professional tolerance limits. Validation
using a sample dataset of a landed residential unit provided by PolyCam demonstrated comparable, and in fact slightly improved,
precision with an average deviation of ±3.5 mm. These results confirm that the proposed workflow is both reliable and generalizable,
maintaining consistent measurement accuracy across different residential property types.
The findings suggest that point cloud measurement derived from mobile LiDAR can serve as a dependable and transparent
alternative to manual measurement for valuation purposes. While certain limitations remain particularly regarding environmental
control and practitioner familiarity with point cloud tools, the overall workflow demonstrates strong potential for adoption in digital
valuation practice. Future studies involving larger datasets and semi-automated approaches could further validate and refine this
method for broader application in property valuation and geospatial analytics.
VI. Acknowledgment
Authors would like to express our gratitude to the Ministry of Higher Education Malaysia (MOHE) for funding this project under
the Fundamental Research Grant Scheme (FRGS) under VOT number R.J130000.7852.5F685 and R.J130000.7852.5F737
References
1. N. French and L. Gabrielli, Property Valuation: The Five Methods, Routledge, 2018.
2. N. Kaluthanthri and A. Hippola, “Uncertainty in valuation practice: Causes and implications,” Property Management, vol.
41, no. 1, pp. 89–104, 2023.
3. M. Mallinson and N. French, “Uncertainty in property valuation,” Journal of Property Investment & Finance, vol. 18, no.
1, pp. 13–32, 2000.
4. Y. Yi, L. Chen, and X. Zhao, “LiDAR applications in urban mapping,” Remote Sensing, vol. 9, no. 9, p. 942, 2017.
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 643
5. H. Zhao, Q. Zhang, and X. Huang, “Advances in LiDAR-based building modeling,” ISPRS Annals of the Photogrammetry,
Remote Sensing and Spatial Information Sciences, vol. V-2-2020, pp. 351–360, 2020.
[6] H. Yamani, M. Azizan, and S. Noor, “Challenges in integrating 3D data in valuation workflows,” Journal of Valuation
Science, vol. 10, no. 2, pp. 15–28, 2021.
6. R. Boeters, H. Li, and S. Zlatanova, “3D models and valuation accuracy: An LOD2 approach,” ISPRS Annals of the
Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. II-3/W4, pp. 31–38, 2015.
7. A. Ganz, N. Haala, and G. Mandlburger, “Accuracy analysis of handheld LiDAR devices for indoor mapping,” ISPRS
Archives, vol. XLII-2/W13, pp. 633–640, 2019.
8. M. Arif, S. Rahman, and H. Zulkifli, “Application of consumer-grade LiDAR for built environment measurement,” Journal
of Spatial Technologies, vol. 12, no. 3, pp. 44–52, 2024.
9. L. Cai, Y. Wang, and X. Zhou, “Satellite-based LiDAR for urban building measurement: Accuracy and limitations,” ISPRS
Journal of Photogrammetry and Remote Sensing, vol. 212, pp. 112–124, 2024.
10. M. Borz, D. Rusu, and R. Muresan, “Evaluation of mobile LiDAR accuracy in built environment applications,” Remote
Sensing Applications, vol. 18, no. 1, p. 101112, 2024.
11. Board of Valuers, Appraisers, Estate Agents and Property Managers (BOVAEP), Malaysian Valuation Standards (8th
Edition), Putrajaya: Ministry of Finance Malaysia, 2022.
12. B. Mete and T. Yomralioglu, “Integration of BIM and GIS for 3D property valuation,” Land Use Policy, vol. 82, pp. 524–
532, 2019.
13. B. Mete, I. Turan, and T. Yomralioglu, “GIS-BIM integration for automated valuation models,” Computers, Environment
and Urban Systems, vol. 96, p. 101222, 2022.
14. B. Atazadeh, A. Rajabifard, M. Kalantari, and K. Champion, “Extending a BIM-based data model to support 3D digital
management of complex ownership spaces,” International Journal of Geographical Information Science, vol. 31, no. 7, pp.
1440–1463, 2017.
15. M. Renigier-Biłozor, S. Źróbek, and R. Źróbek, “Integration of uncertainty in automated property valuation,” Land Use
Policy, vol. 82, pp. 723–735, 2019.
16. Y. Su, J. Zhang, and Y. Zhao, “Automated real estate valuation using BIM and machine learning,” Automation in
Construction, vol. 126, p. 103116, 202