Research map / 3D vision / Gaussian splatting
TG-Field: Geometry-Aware Radiative Gaussian Fields for Tomographic Reconstruction
Yuxiang Zhong, Jun Wei, Chaoqi Chen, Senyou An, Hui Huang
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 semantic-consistency regularization (ablation) | TG-Field hash encoder, static CT · Static CT ablation | PSNR: 28.71 -> 28.95 | Improved |
| add semantic-consistency regularization (ablation) | TG-Field hash encoder, static CT · Static CT ablation | SSIM: 0.841 -> 0.849 | Improved |
| add spatiotemporal attention block (ablation) | TG-Field HE, dynamic · 4D CT ablation | PSNR: 34.26 -> 34.89 | Improved |
| add spatiotemporal attention block (ablation) | TG-Field HE, dynamic · 4D CT ablation | SSIM: 0.939 -> 0.945 | Improved |
| add motion-flow module (ablation) | TG-Field HE+STAB, dynamic · 4D CT ablation | PSNR: 34.89 -> 35.23 | Improved |
| add motion-flow module (ablation) | TG-Field HE+STAB, dynamic · 4D CT ablation | SSIM: 0.945 -> 0.952 | Improved |
| add semantic regularization (ablation) | TG-Field HE+STAB+MF, dynamic · 4D CT ablation | PSNR: 35.23 -> 35.41 | Improved |
| add semantic regularization (ablation) | TG-Field HE+STAB+MF, dynamic · 4D CT ablation | SSIM: 0.952 -> 0.955 | Improved |
| use iteration-based (CGLS + ASD-POCS) initialization instead of cubic uniform or FDK initialization (ablation) | TG-Field · Object CT, 2-8 views | PSNR | Improved |
| replace R2-Gaussian with TG-Field geometry-aware deformation | R2-Gaussian · Synthetic static CT (5 views) | PSNR: 23.81 -> 24.54 | Improved |
| replace R2-Gaussian with TG-Field | R2-Gaussian · Real static CT (10 views) | PSNR: 32.73 -> 33.59 | Improved |
| replace 4DGS with TG-Field | 4DGS · 4D CT average | PSNR: 32.80 -> 33.78 | Improved |
Closest papers by meaning
- R$^2$-Gaussian: Rectifying Radiative Gaussian Splatting for Tomographic Reconstruction (NeurIPS 2024 · similarity 0.91)
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.88)
- Gaussian-Flow: 4D Reconstruction with Dynamic 3D Gaussian Particle (CVPR 2024 · similarity 0.88)
- Volumetrically Consistent 3D Gaussian Rasterization (CVPR 2025 · similarity 0.88)
- Learning Efficient Fuse-and-Refine for Feed-Forward 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.88)
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.
