Multiview 2D/3D Rigid Registration via a Point-Of-Interest Network for Tracking and Triangulation
Abstract
We propose to tackle the problem of multiview 2D/3D rigid registration for intervention via a Point-Of-Interest Network for Tracking and Triangulation (POINT2). POINT2learns to establish 2D point-to-point correspondences between the pre- and intra-intervention images by tracking a set of random POIs. The 3D pose of the pre-intervention volume is then estimated through a triangulation layer. In POINT2, the unified framework of the POI tracker and the triangulation layer enables learning informative 2D features and estimating 3D pose jointly. In contrast to existing approaches, POINT2only requires a single forward-pass to achieve a reliable 2D/3D registration. As the POI tracker is shift-invariant, POINT2is more robust to the initial pose of the 3D pre-intervention image. Extensive experiments on a large-scale clinical cone-beam CT (CBCT) dataset show that the proposed POINT2method outperforms the existing learning-based method in terms of accuracy, robustness and running time. Furthermore, when used as an initial pose estimator, our method also improves the robustness and speed of the state-of-the-art optimization-based approaches by ten folds.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Despite matching random points, Liao et al.'s method is likely still scene-specific because matching is performed using a very deep CNN with very large receptive field which may have learned to exploit the global context of the random keypoints, which would not be preserved in different anatomy.
“However, despite matching random points, this method is likely still scene-specific because matching is performed using a very deep CNN with very large receptive field which may have learned to exploit the global context of the random keypoints, which would not be preserved in different anatomy.”
From The Impact of Machine Learning on 2D/3D Registration for Image-Guided Interventions: A Systematic Review and Perspective (2021) · cites this paper as Liao et al., 2019
Liao et al. (2019) method learns to establish sparse point correspondences between random points rather than anatomically relevant ones.
“Another noteworthy method learns to establish sparse point correspondences between random points rather than anatomically relevant ones (Liao et al., 2019).”
From The Impact of Machine Learning on 2D/3D Registration for Image-Guided Interventions: A Systematic Review and Perspective (2021) · cites this paper as Liao et al., 2019
Methods that model 2D/3D correspondences will require an additional dimension to assign points to objects, adding complexity and potentially increasing training challenges.
“Methods which model 2D/3D correspondences (Schaffert et al., 2019, 2020a,b; Liao et al., 2019) will require an additional dimension to handle the assignment of points to various obejcts, adding complexity and potentially increasing the challenges associated with training.”
From The Impact of Machine Learning on 2D/3D Registration for Image-Guided Interventions: A Systematic Review and Perspective (2021) · cites this paper as Liao et al., 2019
No citation lists for this paper
No citation lists for this paper yet — checked again daily.
References 28 · Cited by 66
- Authors
- Liao, H., Lin, W.-A., Zhang, J., Zhang, J., Luo, J., & Zhou, S. K.
- Affiliations
- University of Rochester (US)
- University of Maryland, College Park (US)
- Rutgers, The State University of New Jersey (US)
- Chinese Academy of Sciences (CN)
- Venue
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2019-06
- Pages
- 12630-12639
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2963332432
- Cited by
- 66
- References
- 28
- Citations by year
- 2026: 3 · 2025: 8 · 2024: 16 · 2023: 5 · 2022: 8 · 2021: 13 · 2020: 11 · 2019: 2
- FWCI
- 2.77
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Health Sciences › Dentistry › Oral Surgery › Dental Radiography and Imaging
- Physical Sciences › Engineering › Computational Mechanics › 3D Shape Modeling and Analysis
- Keywords
- triangulation, point-to-point correspondence, cone-beam computed tomography, 3D pose estimation, initial pose estimation, medical image registration
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
