loading page

Assessing the Practical Applicability of Neural-Based Point Clouds Registration Algorithms: A Comparative Analysis
  • Simone Fontana,
  • Federica Di Lauro,
  • Domenico G. Sorrenti
Simone Fontana
Universita degli Studi di Milano-Bicocca Dipartimento di Giurisprudenza

Corresponding Author:[email protected]

Author Profile
Federica Di Lauro
Universita degli Studi di Milano-Bicocca Dipartimento di Informatica Sistemistica e Comunicazione
Author Profile
Domenico G. Sorrenti
Universita degli Studi di Milano-Bicocca Dipartimento di Informatica Sistemistica e Comunicazione
Author Profile


Point cloud registration is a vital task in 3D perception, with several different applications in robotics. Recent advancements have introduced neural-based techniques that promise enhanced accuracy and robustness. In this paper, we thoroughly evaluate well-known neural-based point cloud registration methods using the Point Clouds Registration Benchmark, which was developed to cover a large variety of use cases. Our evaluation focuses on the performance of these techniques when applied to real-complex data, which presents a more challenging and realistic scenario than the simpler experiments typically conducted by the original authors. With few exceptions, the results reveal a significant performance gap, with most neural-based methods performing poorly on real-complex data. However, amidst the underwhelming results, 3DSmoothNet stands out as an exception, demonstrating excellent performance across the majority of benchmark sequences. Remarkably, it outperforms a state-of-the-art feature extractor, namely FPFH, showcasing its effectiveness in capturing informative and discriminative features from point clouds. Given these results, we assert that while neural-based techniques could have advantages over conventional methods, further research is necessary to fully unlock their practical utility. We advocate for more extensive studies employing unbiased and realistic testing procedures, akin to the rigorous evaluation framework employed in this work. This paper represents a crucial step towards establishing a more robust literature in the field of point cloud registration. By exposing the limitations and advantages of existing neural-based methods, we aim to inspire future research and drive the development of more effective techniques for real-world applications.
11 Jul 2023Submitted to Journal of Field Robotics
11 Jul 2023Submission Checks Completed
11 Jul 2023Assigned to Editor
12 Jul 2023Review(s) Completed, Editorial Evaluation Pending
10 Aug 2023Reviewer(s) Assigned
04 Oct 2023Editorial Decision: Revise Major
24 Jan 2024Review(s) Completed, Editorial Evaluation Pending
28 Jan 2024Editorial Decision: Revise Major
05 Feb 20242nd Revision Received
05 Feb 2024Review(s) Completed, Editorial Evaluation Pending
05 Feb 2024Submission Checks Completed
05 Feb 2024Assigned to Editor