A Survey of Non‐Rigid 3D Registration

Deng, Bailin; Yao, Yuhang; Dyke, Roberto M.; Zhang, Juyong

doi:10.1111/cgf.14502

Cited by 55 publications

(18 citation statements)

References 303 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shape Matching Shape matching is a key problem in 3D shape analysis and has been extensively studied in recent decades [37,87,9,38,13,72,22]. Earlier works focused primarily on hand-crafted 3D local features for matching, including both extrinsic [47,33,75,74,76] and intrinsic [81,4] descriptors.…”

Section: Related Workmentioning

confidence: 99%

“…1 (a). Thus, so far, there is limited success in training local features purely with contrastive learning for direct nearest-neighbor matching, e.g., for non-rigid shape correspondence [22]. Motivated by the above discussion, we introduce a novel smoothness-regularized contrastive learning approach, enabling robust feature-based matching of deformable objects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SRFeat: Learning Locally Accurate and Globally Consistent Non-Rigid Shape Correspondence

Li,

Attaiki,

Ovsjanikov

2022

Preprint

View full text Add to dashboard Cite

In this work, we present a novel learning-based framework that combines the local accuracy of contrastive learning with the global consistency of geometric approaches, for robust non-rigid matching. We first observe that while contrastive learning can lead to powerful point-wise features, the learned correspondences commonly lack smoothness and consistency, owing to the purely combinatorial nature of the standard contrastive losses. To overcome this limitation we propose to boost contrastive feature learning with two types of smoothness regularization that inject geometric information into correspondence learning. With this novel combination in hand, the resulting features are both highly discriminative across individual points, and, at the same time, lead to robust and consistent correspondences, through simple proximity queries. Our framework is general and is applicable to local feature learning in both the 3D and 2D domains. We demonstrate the superiority of our approach through extensive experiments on a wide range of challenging matching benchmarks, including 3D non-rigid shape correspondence and 2D image keypoint matching.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

SRFeat: Learning Locally Accurate and Globally Consistent Non-Rigid Shape Correspondence

Li,

Attaiki,

Ovsjanikov

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In the following paragraphs, we review the works that are most closely related to our approach. A more complete overview can be found in recent surveys [17,18], and more recently [4] (Section 4).…”

Section: Related Workmentioning

confidence: 99%

“…The algorithmic challenge of robust shape matching primarily lies in the fact that shapes may undergo significant variations, such as arbitrary non-rigid deformations. Earlier works to tackle non-rigid shape correspondence conventionally build upon hand-crafted features and pipelines [3], while with the advent of deep learning, the research focus has largely shifted to data-driven and learning-based approaches for improved matching robustness and accuracy [4].…”

Section: Introductionmentioning

confidence: 99%

Learning Multi-resolution Functional Maps with Spectral Attention for Robust Shape Matching

Li,

Donati,

Ovsjanikov

2022

Preprint

View full text Add to dashboard Cite

In this work, we present a novel non-rigid shape matching framework based on multi-resolution functional maps with spectral attention. Existing functional map learning methods all rely on the critical choice of the spectral resolution hyperparameter, which can severely affect the overall accuracy or lead to overfitting, if not chosen carefully. In this paper, we show that spectral resolution tuning can be alleviated by introducing spectral attention. Our framework is applicable in both supervised and unsupervised settings, and we show that it is possible to train the network so that it can adapt the spectral resolution, depending on the given shape input. More specifically, we propose to compute multi-resolution functional maps that characterize correspondence across a range of spectral resolutions, and introduce a spectral attention network that helps to combine this representation into a single coherent final correspondence. Our approach is not only accurate with nearisometric input, for which a high spectral resolution is typically preferred, but also robust and able to produce reasonable matching even in the presence of significant non-isometric distortion, which poses great challenges to existing methods. We demonstrate the superior performance of our approach through experiments on a suite of challenging near-isometric and non-isometric shape matching benchmarks.

show abstract

“…Non-rigid point cloud registration [1], which is the surface registration of multiple non-rigid 3D point clouds optimally together into a common coordinate system, is important for 3D vision and graphics related areas, such as medical imaging, virtual reality and robotics. The widespread usage of low-cost 3D scanning devices, such as Microsoft Kinect, further promotes it.…”

Section: Introductionmentioning

confidence: 99%

Automatic Extraction of the Sparse Prior Correspondences for Non-Rigid Point Cloud Registration

Zhu¹,

Tian²,

Ye³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Non-rigid registration of point clouds is still far from stable, especially for the largely deformed one. Sparse initial correspondences are often adopted to facilitate the process. However, there are few studies on how to build them automatically. Therefore, in this paper, we propose a robust method to compute such priors automatically, where a global and local combined strategy is adopted. These priors in different degrees of deformation are obtained by the locally geometrical-consistent point matches from the globally structural-consistent region correspondences. To further utilize the matches, this paper also proposes a novel registration method based on the Coherent Point Drift framework. This method takes both the spatial proximity and local structural consistency of the priors as supervision of the registration process and thus obtains a robust alignment for clouds with significantly different deformations. Qualitative and quantitative experiments demonstrate the advantages of the proposed method.

show abstract

A Survey of Non‐Rigid 3D Registration

Cited by 55 publications

References 303 publications

SRFeat: Learning Locally Accurate and Globally Consistent Non-Rigid Shape Correspondence

SRFeat: Learning Locally Accurate and Globally Consistent Non-Rigid Shape Correspondence

Learning Multi-resolution Functional Maps with Spectral Attention for Robust Shape Matching

Automatic Extraction of the Sparse Prior Correspondences for Non-Rigid Point Cloud Registration

Contact Info

Product

Resources

About