A benchmark for point clouds registration algorithms

Fontana, Simone; Cattaneo, Daniele; Ballardini, Augusto Luis; Vaghi, Matteo; Sorrenti, Domenico G.

doi:10.1016/j.robot.2021.103734

Cited by 28 publications

(37 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A notable one is generalized ICP (GICP) [ 14 ], which incorporates the covariance of the points into the standard ICP error function and, by doing so, obtains much better results [ 40 ], although at the expense of computational time.…”

Section: Related Workmentioning

confidence: 99%

“…To align the two PCs, we used generalized ICP [ 14 ] because, among the many different solutions to local point clouds registration, it provides very good results with a reasonable computational time [ 39 , 40 ]. Although recently new approaches to point clouds registration have been introduced, such as Teaser++ [ 50 ] and fast global registration [ 49 ], they are mainly aimed at global registration and have not been proven to outperform the best local registration algorithms, such as GICP.…”

Section: Proposed Localization Pipelinementioning

confidence: 99%

See 1 more Smart Citation

Vehicle Localization Using 3D Building Models and Point Cloud Matching

Ballardini

Fontana

Cattaneo

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Detecting buildings in the surroundings of an urban vehicle and matching them to building models available on map services is an emerging trend in robotics localization for urban vehicles. In this paper, we present a novel technique, which improves a previous work by detecting building façade, their positions, and finding the correspondences with their 3D models, available in OpenStreetMap. The proposed technique uses segmented point clouds produced using stereo images, processed by a convolutional neural network. The point clouds of the façades are then matched against a reference point cloud, produced extruding the buildings’ outlines, which are available on OpenStreetMap (OSM). In order to produce a lane-level localization of the vehicle, the resulting information is then fed into our probabilistic framework, called Road Layout Estimation (RLE). We prove the effectiveness of this proposal, testing it on sequences from the well-known KITTI dataset and comparing the results concerning a basic RLE version without the proposed pipeline.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Proposed Localization Pipelinementioning

confidence: 99%

Vehicle Localization Using 3D Building Models and Point Cloud Matching

Ballardini

Fontana

Cattaneo

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…This dataset is considered to be difficult to work with because of complications related to noise and irregular density. We will fine-tune our network on ETH dataset [30], compare with other methods, and use the dataset prepared by Fontana et al [16] to evaluate our results. For ETH, the evaluation protocol that we have chosen is different from previous papers.…”

Section: Eth Dataset [30] (Target Dataset)mentioning

confidence: 99%

“…For ETH, the evaluation protocol that we have chosen is different from previous papers. We prefer using the more rigorous protocol described by Fontana et al [16]: we compare our method on the 8 scenes of the dataset and not just the 4 included in previous published works like [4,14,2,29,19].…”

Section: Eth Dataset [30] (Target Dataset)mentioning

confidence: 99%

See 1 more Smart Citation

3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning

Horache¹,

Deschaud²,

Goulette³

2021

Preprint

View full text Add to dashboard Cite

We present MS-SVConv, a fast multi-scale deep neural network that outputs features from point clouds for 3D registration between two scenes. We compute features using a 3D sparse voxel convolutional network on a point cloud at different scales and then fuse the features through fullyconnected layers. With supervised learning, we show significant improvements compared to state-of-the-art methods on the competitive and well-known 3DMatch benchmark. We also achieve a better generalization through different source and target datasets, with very fast computation. Finally, we present a strategy to fine-tune MS-SVConv on unknown datasets in a self-supervised way, which leads to state-of-the-art results on ETH and TUM datasets. The code is publicly available at: https://github.com/ humanpose1/MS-SVConv.

show abstract

Assessing the practical applicability of neural‐based point clouds registration algorithms: A comparative analysis

Fontana,

Di Lauro,

Sorrenti

2024

Journal of Field Robotics

View full text Add to dashboard Cite

Point cloud registration is a vital task in three‐dimensional (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. The results reveal considerable variability in performance across different techniques, highlighting the importance of assessing algorithms in realistic settings. Notably, 3DSmoothNet emerges as a standout solution, demonstrating good and consistent results across various data sets. Its efficacy, coupled with a relatively low graphics processing unit (GPU) memory footprint, makes it a promising choice for robotics applications, even if it is not yet suitable for real‐time applications due to its execution time. Fully Convolutional Geometric Features also performs well, albeit with greater variability among data sets. PREDATOR and GeoTransformer are promising, but demand substantial GPU memory, when handling large point clouds from the Point Clouds Registration Benchmark. A notable finding concerns the performance of Fast Point Feature Histograms, which exhibit results comparable to the best approaches while demanding minimal computational resources. Overall, this comparative analysis provides valuable insights into the strengths and limitations of neural‐based registration techniques, both in terms of the quality of the results and the computational resources required. This helps researchers to make informed decisions for robotics applications.

show abstract

A benchmark for point clouds registration algorithms

Cited by 28 publications

References 19 publications

Vehicle Localization Using 3D Building Models and Point Cloud Matching

Vehicle Localization Using 3D Building Models and Point Cloud Matching

3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning

Assessing the practical applicability of neural‐based point clouds registration algorithms: A comparative analysis

Contact Info

Product

Resources

About