Research map / Vision-language and multimodal / Cross-modal learning
Unsupervised Multi-modal Medical Image Registration via Invertible Translation
Mengjie Guo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add INN cross-translation loss to base model (ablation) | INNReg base · T1/T2 MRI | Dice (%): 83.026 -> 83.469 | Improved |
| add plain NMI loss on top of cross translation (ablation) | INNReg base · T1/T2 MRI | Dice (%): 83.469 -> 82.153 | Hurt |
| use barrier NMI loss instead of plain NMI (ablation) | INNReg · T1/T2 MRI | Dice (%): 82.153 -> 86.84 | Improved |
| add DDWConv local attention to translation network (ablation) | INNReg · T1/T2 MRI | HD95: 136.84976 -> 101.37179 | Improved |
| add DDWConv local attention to translation network (ablation) | INNReg · T1/T2 MRI | SSIM (%): 90.84 -> 90.555 | Hurt |
| replace GAN translation with invertible network translation | translation-based registration · MRI/CT | SSIM (%): 71.637 -> 73.067 | Improved |
| INNReg versus Arar et al. on deformation smoothness | INNReg · T1/T2 MRI | non-positive Jacobian (%): 0.007 vs 0.101 | Hurt |
Closest papers by meaning
- Unsupervised Multi-Modal Image Registration via Geometry Preserving Image-to-Image Translation (CVPR 2020 · similarity 0.90)
- Foundation-Guided Representation Alignment for Multimodal Medical Image Registration (ECCV 2026 · similarity 0.88)
- JSSR: A Joint Synthesis, Segmentation, and Registration System for 3D Multi-Modal Image Alignment of Large-scale Pathological CT Scans (ECCV 2020 · similarity 0.86)
- Breaking the Dilemma of Medical Image-to-image Translation (NeurIPS 2021 · similarity 0.86)
- Multi-modal Vision Pre-training for Medical Image Analysis (CVPR 2025 · similarity 0.86)
- Modality-Agnostic Structural Image Representation Learning for Deformable Multi-Modality Medical Image Registration (CVPR 2024 · similarity 0.85)
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.
