Research map / 3D vision / Gaussian splatting
DentalGS: Pose-Free 3D Gaussian Splatting from Five Intraoral Images for Novel View Synthesis
Honghao Dai, Yuanfeng Zhou, Guangshun Wei, Zhihao Li, Wenping Wang
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 perceptual loss on restored images (ablation) | DentalGS · Intraoral dataset (5-view) | LPIPS: 0.2770 -> 0.2470 | Improved |
| add structure loss (ablation) | DentalGS · Intraoral dataset (5-view) | PSNR: 24.45 -> 24.54 | Improved |
| add SSIM loss (ablation) | DentalGS · Intraoral dataset (5-view) | PSNR: 24.54 -> 24.84 | Improved |
| use losses on low-quality restored images versus high-quality ones (ablation) | DentalGS · Intraoral dataset (5-view) | SSIM | Mixed |
| replace UNet RepairNet restoration with SDM restoration (GSOBJ) (ablation) | GSOBJ · Intraoral dataset (5-view) | PSNR: 21.43 -> 20.91 | Hurt |
| increase input views from 5 to 7 | DentalGS · Intraoral dataset | PSNR: 26.20 -> 26.86 | No clear effect |
| increase input views from 5 to 7 | 3DGS · Intraoral dataset | PSNR: 19.71 -> 24.02 | Improved |
| use DentalGS instead of GSOBJ (5-view) | DentalGS · Intraoral dataset (5-view) | SSIM: 0.8044 -> 0.9358 | Improved |
| use DentalGS instead of GSOBJ (5-view) | DentalGS · Intraoral dataset (5-view) | LPIPS: 0.2438 -> 0.2689 | Hurt |
Closest papers by meaning
- Teeth-SEG: An Efficient Instance Segmentation Framework for Orthodontic Treatment based on Multi-Scale Aggregation and Anthropic Prior Knowledge (CVPR 2024 · similarity 0.85)
- Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images (CVPR 2025 · similarity 0.85)
- OralGPT-Plus: Learning to Use Visual Tools via Reinforcement Learning for Panoramic X-ray Analysis (CVPR 2026 · similarity 0.84)
- Volumetrically Consistent 3D Gaussian Rasterization (CVPR 2025 · similarity 0.84)
- NeBLa: Neural Beer-Lambert for 3D Reconstruction of Oral Structures from Panoramic Radiographs (AAAI 2024 · similarity 0.84)
- Collaborative Tooth Motion Diffusion Model in Digital Orthodontics (AAAI 2024 · similarity 0.84)
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.
