Comparison of Low-Cost 3D Reconstruction Techniques for Change Detection in Indoor Environments

Walter Bueno de Brito Neto; Michel M. Silva; Thiago L. Gomes

doi:10.22456/2175-2745.150907

Authors

Walter Bueno de Brito Neto Universidade Federal de Viçosa (UFV)
Michel M. Silva Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0002-2499-9619
Thiago L. Gomes Universidade Federal de Viçosa (UFV)

DOI:

https://doi.org/10.22456/2175-2745.150907

Keywords:

3D reconstruction, change detection, point clouds, deep learning

Abstract

This work evaluates low-cost 3D capture techniques for change detection in indoor environments. We compared a traditional Structure-from-Motion method (COLMAP) with a deep learning approach (CUT3R), using RGB images captured in two sessions of the same laboratory, one with visual changes and one without. The pipeline included 3D reconstruction, point cloud post-processing, alignment, and point-to-point (Cloud-to-Cloud) change detection. Results show that CUT3R produces visually richer reconstructions with significantly lower processing time but suffers from scale inconsistencies and object duplication, compromising precise change detection. COLMAP, although slower, maintains spatial coherence and provides more reliable identification of alterations. These findings indicate that traditional methods are still preferable for rigorous change detection, while deep learning approaches are advantageous for scenarios requiring fast or visually detailed reconstructions.

Downloads

Download data is not yet available.

References

[1] QIN, R. 3D change detection in an urban environment with multi-temporal data. ETH Zurich, 2015.

[2] KHARROUBI, A. et al. Three-dimensional change detection using point clouds: A review. Geomatics, v. 2, n. 4, p. 457–485, 2022. ISSN 2673-7418. Disponível em: https://www.mdpi.com/2673-7418/2/4/25.

[3] XIAO, W. et al. 3D urban object change detection from aerial and terrestrial point clouds: A review. International Journal of Applied Earth Observation and Geoinformation, v. 118, p. 103258, 2023. ISSN 1569-8432. Disponível em: https://www.sciencedirect.com/science/article/pii/S1569843223000808.

[4] TANEJA, A.; BALLAN, L.; POLLEFEYS, M. Geometric change detection in urban environments using images. IEEE Transactions on Pattern Analysis and Machine Intelligence, v. 37, p. 2193–2206, 2015.

[5] VIEIRA, A.; DREWS-JR, P.; CAMPOS, M. Spatial density patterns for efficient change detection in 3D environment for autonomous surveillance robots. IEEE Transactions on Automation Science and Engineering, v. 11, p. 766–774, 2014.

[6] TINOCO, C. R.; TRAVENCOLO, B. A. N.; MARTINS, L. G. A. Evaluation of a coordination model for robot swarms in collective environment mapping. In: 2025 IEEE Congress on Evolutionary Computation (CEC). [S.l.: s.n.], 2025. p. 1–8.

[7] GÖNCZ, L.; MAJDIK, A. Object-based change detection algorithm with a spatial AI stereo camera. Sensors (Basel, Switzerland), v. 22, 2022.

[8] COLLIS, R. Lidar. Applied Optics, v. 9, p. 1782–1788, 2019.

[9] KESELMAN, L. et al. Intel RealSense stereoscopic depth cameras. CoRR, abs/1705.05548, 2017. Disponível em: http://arxiv.org/abs/1705.05548.

[10] ULLMAN, S. The interpretation of structure from motion. Proceedings of the Royal Society of London. Series B. Biological Sciences, v. 203, p. 405–426, 1979.

[11] FURUKAWA, Y.; HERNÁNDEZ, C. Multi-view stereo: A tutorial. Found. Trends Comput. Graph. Vis., v. 9, p. 1–148, 2015.

[12] GOMES, L.; REGINA PEREIRA BELLON, O.; SILVA, L. 3D reconstruction methods for digital preservation of cultural heritage: A survey. Pattern Recognition Letters, v. 50, p. 3–14, 2014. ISSN 0167-8655. Depth Image Analysis. Disponível em: https://www.sciencedirect.com/science/article/pii/S0167865514001032.

[13] LUO, Z. et al. The NeRF signature: Codebook-aided watermarking for neural radiance fields. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025.

[14] DOSOVITSKIY, A. et al. An image is worth 16x16 words: Transformers for image recognition at scale. 2021. Disponível em: https://arxiv.org/abs/2010.11929.

[15] WANG, Q. et al. Continuous 3D perception model with persistent state. 2025. Disponível em: https://arxiv.org/abs/2501.12387.

[16] SINGH, A. Review article: Digital change detection techniques using remotely-sensed data. International Journal of Remote Sensing, Taylor & Francis, v. 10, n. 6, p. 989–1003, 1989. Disponível em: https://doi.org/10.1080/01431168908903939.

[17] QIN, R.; TIAN, J.; REINARTZ, P. 3D change detection – approaches and applications. ISPRS Journal of Photogrammetry and Remote Sensing, v. 122, p. 41–56, 2016. ISSN 0924-2716. Disponível em: https://www.sciencedirect.com/science/article/pii/S0924271616304026.

[18] VERYKOKOU, S.; IOANNIDIS, C. An overview on image-based and scanner-based 3D modeling technologies. Sensors, Basel, Switzerland, v. 23, n. 2, p. 596, jan 2023.

[19] SCHÖNBERGER, J. L. et al. A vote-and-verify strategy for fast spatial verification in image retrieval. In: Asian Conference on Computer Vision (ACCV). [S.l.: s.n.], 2016.

[20] SAMAVATI, T.; SORYANI, M. Deep learning-based 3D reconstruction: a survey. Artif. Intell. Rev., Kluwer Academic Publishers, USA, v. 56, n. 9, p. 9175–9219, jan. 2023. ISSN 0269-2821. Disponível em: https://doi.org/10.1007/s10462-023-10399-2.

[21] MANDIKAL, P. et al. 3D-LMNet: Latent embedding matching for accurate and diverse 3D point cloud reconstruction from a single image. CoRR, abs/1807.07796, 2018. Disponível em: http://arxiv.org/abs/1807.07796.