Research map / 3D vision / Gaussian splatting
Dynamic CT Reconstruction From Limited Views With Implicit Neural Representations and Parametric Motion Fields
Albert W. Reed, Hyojin Kim, Rushil Anirudh, K. Aditya Mohan, Kyle Champley, Jingu Kang, Suren Jayasuriya
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 |
|---|---|---|---|
| use INR with parametric motion field instead of TIMBIR on limited-angle 4D CT | Ours · D4DCT Alum#1 | PSNR (dB): 10.95 -> 22.68 | Improved |
| use INR with parametric motion field instead of Warp and Project | Ours · Thoracic CT | PSNR (dB): 8.27 -> 22.45 | Improved |
| add coarse-to-fine hierarchical motion model (ablation) | Ours · D4DCT aluminum objects | PSNR (dB): 6 dB performance improvement | Improved |
| add polynomial order 3 or higher in warp field (vs low order) (ablation) | Ours · D4DCT aluminum objects | reconstruction quality | Improved |
| query INR at 256^3 instead of trilinear upsampling of the 80^3 reconstruction (ablation) | Ours · Thoracic CT upsampling (3D coronal view) | PSNR (dB): 25.27 -> 25.44 | Improved |
| query INR at 256^3 instead of trilinear upsampling (2D coronal view) (ablation) | Ours · Thoracic CT upsampling (2D coronal view) | PSNR (dB): 19.70 -> 20.26 | Improved |
| use INR instead of SART/FBP for 3D Shepp-Logan limited view reconstruction (ablation) | static 3D reconstruction · 3D Shepp-Logan | PSNR | Improved |
| widen angular range from 180 to 720 degrees at 90 projections (ablation) | Ours · D4DCT Alum#5 | PSNR (dB): 25.30 -> 27.69 | Improved |
| widen angular range from 180 to 720 degrees at 90 projections (ablation) | TIMBIR · D4DCT Alum#5 | PSNR (dB): 11.20 -> 16.28 | Improved |
Closest papers by meaning
- Spatiotemporal-Aware Neural Fields for Dynamic CT Reconstruction (AAAI 2025 · similarity 0.91)
- Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion (CVPR 2018 · similarity 0.88)
- MoRe: Motion-aware Feed-forward 4D Reconstruction Transformer (CVPR 2026 · similarity 0.87)
- Neuromorphic X-ray Computed Tomography (ECCV 2026 · similarity 0.87)
- ReMatching Dynamic Reconstruction Flow (ICLR 2025 · similarity 0.85)
- C^2RV: Cross-Regional and Cross-View Learning for Sparse-View CBCT Reconstruction (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.
