Research map / 3D vision / Gaussian splatting
MedGSSR: Generalizable Medical Image Super-Resolution 3D Reconstruction via Hierarchical Feed-forward Gaussian Splatting
Chengkai Wang, Luoyu Hong, Yiting Zhao, Jiamin Wang, Xiang Feng, Feiwei Qin, Zhenzhong Kuang, Xuefei Yin, Ali Bashashati, Yanming Zhu
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 fine branch to coarse-only model (structure-texture decoupling) (ablation) | MedGSSR · HCP MRI 4x | PSNR: 35.24 -> 36.60 | Improved |
| add fine branch to coarse-only model (ablation) | MedGSSR · HCP MRI 4x | LPIPS: 0.0789 -> 0.0363 | Improved |
| increase sub-voxel count from 1 to 4 (ablation) | MedGSSR · HCP MRI 4x | PSNR: 35.98 -> 36.60 | Improved |
| increase sub-voxel count from 4 to 8 (ablation) | MedGSSR · HCP MRI 4x | PSNR: 36.60 -> 36.52 | No clear effect |
| increase sub-voxel count from 1 to 4 (ablation) | MedGSSR · MELA CT 4x | PSNR: 36.42 -> 37.31 | Improved |
| add FFT loss (ablation) | MedGSSR · HCP cross-domain 4x | LPIPS: 0.1253 -> 0.1078 | Mixed |
| add FFT loss (ablation) | MedGSSR · HCP cross-domain 4x | PSNR: 35.84 -> 35.72 | Hurt |
| use full truncation radius instead of 1 sigma (ablation) | MedGSSR · HCP cross-domain 4x | PSNR: 31.05 -> 35.84 | Improved |
| use full truncation radius instead of 1 sigma (ablation) | MedGSSR · MELA CT 4x | PSNR: 33.47 -> 37.31 | Improved |
| train on full instead of 50% data (ablation) | MedGSSR · HCP cross-domain 4x | PSNR: 34.26 -> 35.84 | Improved |
| train on full instead of 75% data (ablation) | MedGSSR · HCP cross-domain 4x | PSNR: 35.49 -> 35.84 | Improved |
| feed-forward Gaussian method vs ArSSR | MedGSSR · MSD MRI 4x | PSNR: 28.51 -> 32.10 | Improved |
Closest papers by meaning
- TG-Field: Geometry-Aware Radiative Gaussian Fields for Tomographic Reconstruction (AAAI 2026 · similarity 0.88)
- Adaptive Anisotropic Gaussian Splatting for Multi-contrast MRI Arbitrary-Scale Super-Resolution with Anatomy Guidance (CVPR 2026 · similarity 0.88)
- R$^2$-Gaussian: Rectifying Radiative Gaussian Splatting for Tomographic Reconstruction (NeurIPS 2024 · similarity 0.87)
- CycleINR: Cycle Implicit Neural Representation for Arbitrary-Scale Volumetric Super-Resolution of Medical Data (CVPR 2024 · similarity 0.87)
- RoGSplat: Learning Robust Generalizable Human Gaussian Splatting from Sparse Multi-View Images (CVPR 2025 · similarity 0.86)
- S2Gaussian: Sparse-View Super-Resolution 3D Gaussian Splatting (CVPR 2025 · similarity 0.86)
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.
