Accurate Registration of Cross-Modality Geometry via Consistent Clustering

Zhao, Mingyang; Huang, Xiaoshui; Jiang, Jingen; Mou, Luntian; Yan, Dong‐Ming; Ма, Лей

doi:10.1109/tvcg.2023.3247169

Cited by 9 publications

(4 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recall TE RE ICP [8] 24.3 0.38 5.71 RANSAC [32] 3.47 0.13 8.30 TrICP [9] 7.92 0.18 6.40 CICP [10] 2.48 0.28 8.28 PICP [33] 4.45 0.29 10.85 Super4PCS [13] 6.93 0.24 6.38 GCTR [18] 0.50 0.17 7.46 GMM [14] 9.41 0.18 7.92 FilterReg [15] 30.96 0.10 2.45 DGR [23] 36.60 0.04 4.26 IDAM [34] 1.98 0.13 11.37 FMR [24] 17.80 0.10 4.66 GCC [31] 40…”

Section: Methodsmentioning

confidence: 99%

“…The registration recall rate is calculated as the ratio of the number of point cloud pairs with Rotation Error (RE) less than 15 • and a Translation Error (TE) less than 0.3m to the total number of pairs. Consistent with some literature, such as [31], when evaluating average performance, we only consider point cloud pairs that were successfully recalled. This is due to the fact that pairs that fail to be recalled deviate significantly from the baseline data, rendering their performance metrics potentially unreliable.…”

Section: Performancementioning

confidence: 99%

“…Among deep neural network-based methods, apart from IDAM [34], DGR [23], and FMR [24] all exhibit higher-than-average recall rates and relatively lower estimation errors compared to conventional optimization methods. It's noteworthy that the avant-garde cross-modality point cloud registration technique, GCC [31], eclipses the DGR method, which harnesses deep neural network feature mapping, in recall metrics, setting a new benchmark. Nonetheless, its sub-50% recall rate underscores the inherent challenges plaguing the domain of cross-modality point cloud registration.…”

Section: Performancementioning

confidence: 99%

See 2 more Smart Citations

Point Cloud Registration Based on Global and Local Feature Fusion

Yue

Yuan

et al. 2022

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Cross-modality point cloud registration is confronted with significant challenges due to inherent differences in modalities between different sensors. We propose a cross-modality point cloud registration framework FF-LOGO: a cross-modality point cloud registration method with feature filtering and local-global optimization. The crossmodality feature correlation filtering module extracts geometric transformation-invariant features from cross-modality point clouds and achieves point selection by feature matching. We also introduce a cross-modality optimization process, including a local adaptive key region aggregation module and a global modality consistency fusion optimization module. Experimental results demonstrate that our two-stage optimization significantly improves the registration accuracy of the feature association and selection module. Our method achieves a substantial increase in recall rate compared to the current state-of-the-art methods on the 3DCSR dataset, improving from 40.59% to 75.74%. Our code will be available at https://github.com/wangmohan17/FFLOGO.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Performancementioning

confidence: 99%

Section: Performancementioning

confidence: 99%

See 1 more Smart Citation

Point Cloud Registration Based on Global and Local Feature Fusion

Yue

Yuan

et al. 2022

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

show abstract

“…R EGISTRATION of point clouds captured by 3D scanner devices is an active research area, whose applications in 3D computer vision and robotics [1] include 3D scene reconstruction [2], simultaneous localization and mapping (SLAM) [3], indoor modeling [4], and virtual reality [5]. Given a source cloud P and target cloud Q, point cloud registration is dedicated to estimating a 6-Degree-of-Freedom (DoF) transformation T = [R|t] ∈ SE(3) to transform P into the target coordinate system of Q [6].…”

Section: Introductionmentioning

confidence: 99%

A Novel Sketch-Based Framework Utilizing Contour Cues for Efficient Point Cloud Registration

Ma,

Wei

2023

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Point cloud registration is a crucial part of 3D computer vision. Existing point cloud registration methods primarily concentrate on utilizing features such as points, lines, and planes, disregarding the valuable contour cues inherent in the scene. In this article, we propose a novel sketch-based framework for point cloud registration that incorporates contour cues to enhance the point cloud registration task. To fully exploit the abundant information provided by contour cues in the scene, the point cloud is first abstracted into a sketch consisting of contour cues obtained through the utilization of planar features, which greatly preserves the inherent contour information. Subsequently, a local contour geometric descriptor is introduced to encode the contour cues in the sketch. Finally, a voting-based Contour Point Pair Feature (CPPF) framework is employed to fuse planar features, local contour geometric features and point pair geometric features, enabling precise estimation of the pose transformation between pairwise point clouds. Extensive experiments conducted on two large-scale outdoor point cloud datasets and two indoor point cloud datasets validate the effectiveness of the proposed sketchbased method. Our proposed method successfully suppresses rotation and translation errors, ultimately achieving state-of-theart performance.

show abstract

Correspondence-Free Non-Rigid Point Set Registration Using Unsupervised Clustering Analysis

Zhao,

Jiang,

et al. 2024

2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Accurate Registration of Cross-Modality Geometry via Consistent Clustering

Cited by 9 publications

References 49 publications

Point Cloud Registration Based on Global and Local Feature Fusion

Point Cloud Registration Based on Global and Local Feature Fusion

A Novel Sketch-Based Framework Utilizing Contour Cues for Efficient Point Cloud Registration

Correspondence-Free Non-Rigid Point Set Registration Using Unsupervised Clustering Analysis

Contact Info

Product

Resources

About