Research map / Architectures and efficiency / CNN design
Affine Medical Image Registration With Coarse-To-Fine Vision Transformer
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use C2FViT instead of the strongest listed CNN baseline | Affine registration · Template-Matching Normalization (MNI152), without CoM initialization | DSC4: 0.67 ± 0.06 -> 0.71 ± 0.06 | Helped |
| Use C2FViT instead of the strongest listed CNN baseline | Affine registration · Atlas-Based Registration (OASIS), without CoM initialization | DSC23: 0.57 ± 0.07 -> 0.64 ± 0.06 | Helped |
| Use C2FViT instead of ConvNet-Affine | Affine registration · Atlas-Based Registration (OASIS train → LPBA test), without CoM initialization | DSC3: 0.36 ± 0.07 -> 0.47 ± 0.04 | Helped |
| Add center-of-mass initialization (ablation) | C2FViT · Template-Matching Normalization (MNI152) | Initial DSC4: 0.14 ± 0.12 -> 0.49 ± 0.11 | Helped |
| Add progressive spatial transformation (ablation) | C2FViT · Atlas-Based Registration (OASIS) | DSC23: 0.62 -> 0.64 | Helped |
| Add center-of-mass initialization (ablation) | C2FViT · Atlas-Based Registration (OASIS) | DSC23: 0.64 -> 0.66 | Helped |
| Add semi-supervised learning (ablation) | C2FViT · Atlas-Based Registration (OASIS) | DSC23: 0.66 -> 0.69 | Helped |
| Use semi-supervised training instead of unsupervised training (ablation) | C2FViT · Atlas-Based Registration (OASIS train → LPBA test), with CoM initialization | DSC3: 0.54 ± 0.03 -> 0.51 ± 0.03 | Hurt |
| Predict decoupled geometric parameters rather than directly estimating the affine matrix (ablation) | C2FViT · Atlas-Based Registration (OASIS) | DSC23: 0.63 ± 0.06 -> 0.64 ± 0.06 | Helped |
Closest papers by meaning
- Correlation-aware Coarse-to-fine MLPs for Deformable Medical Image Registration (CVPR 2024 · similarity 0.86)
- Deep Learning in Medical Image Registration: Magic or Mirage? (NeurIPS 2024 · similarity 0.85)
- Uncertainty Estimation for Pretrained Medical Image Registration Models via Transformation Equivariance (NeurIPS 2026 · similarity 0.85)
- CARL: A Framework for Equivariant Image Registration (CVPR 2025 · similarity 0.85)
- SACB-Net: Spatial-awareness Convolutions for Medical Image Registration (CVPR 2025 · similarity 0.85)
- Intraoperative 2D/3D Image Registration via Differentiable X-ray Rendering (CVPR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.