Research map / 3D vision / 3D generation
Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion
Read the paperCheck 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 |
|---|---|---|---|
| Complete the missing sinogram angles with CTNet and reconstruct the completed sinogram using WLS | Limited-angle CT reconstruction · Checked-in luggage test set | PSNR (dB): 37.3 -> 38.13 | Helped |
| Complete the missing sinogram angles with CTNet and reconstruct the completed sinogram using WLS | Limited-angle CT reconstruction · Checked-in luggage test set | S-PSNR (dB): 32.08 -> 34.11 | Helped |
| Add implicit sinogram completion before reconstruction rather than directly predicting the CT image (ablation) | CTNet-MSE · Checked-in luggage test set | S-PSNR (dB): 33.70 -> 34.11 | Helped |
| Use WLS rather than FBP after sinogram completion (ablation) | CTNet-MSE · Checked-in luggage test set | PSNR (dB): 37.42 -> 38.13 | Helped |
| Use WLS rather than FBP after sinogram completion (ablation) | CTNet-MSE · Checked-in luggage test set | S-PSNR (dB): 33.22 -> 34.11 | Helped |
| Use adversarial loss instead of MSE loss (ablation) | CTNet with WLS sinogram-completion pipeline · Checked-in luggage test set | PSNR (dB): 38.13 -> 38.08 | No clear effect |
| Use adversarial loss instead of MSE loss (ablation) | CTNet with WLS sinogram-completion pipeline · Checked-in luggage test set | S-PSNR (dB): 34.11 -> 34.14 | No clear effect |
| Reduce the available views at test time below the 90° training range (ablation) | CTNet · Checked-in luggage test set, 70°–90° test views | reconstruction quality | Hurt |
| Apply the method to high-frequency scenes with multiple small objects or intricate designs | CTNet · Checked-in luggage test set, complex scenes | reconstruction quality | No clear effect |
Closest papers by meaning
- Near-Exact Recovery for Tomographic Inverse Problems via Deep Learning (ICML 2022 · similarity 0.87)
- C^2RV: Cross-Regional and Cross-View Learning for Sparse-View CBCT Reconstruction (CVPR 2024 · similarity 0.85)
- QN-Mixer: A Quasi-Newton MLP-Mixer Model for Sparse-View CT Reconstruction (CVPR 2024 · similarity 0.84)
- Spatiotemporal-Aware Neural Fields for Dynamic CT Reconstruction (AAAI 2025 · similarity 0.84)
- Unsupervised Self-Prior Embedding Neural Representation for Iterative Sparse-View CT Reconstruction (AAAI 2025 · similarity 0.83)
- DiffusionBlend: Learning 3D Image Prior through Position-aware Diffusion Score Blending for 3D Computed Tomography Reconstruction (NeurIPS 2024 · similarity 0.82)
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.