Research map / Point clouds / Shape and surface completion
TAlignDiff: Automatic Tooth Alignment assisted by Diffusion-based Transformation Learning
Yunbi Liu, Enqi Tang, Shiyu Li, Hui Shuai, Lei Ma, Juncheng Li, Kuai Yu, Shu Lou, Yongchu Pan, Qingshan Liu
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 tooth centroid offset loss (lambda1=0.1) (ablation) | PRN with reconstruction loss · ISICDM 2024 test | TRE: 0.784 -> 0.748 | Improved |
| add DTMD diffusion losses (lambda2=0.01, lambda3=0.1) (ablation) | PRN with centroid loss · ISICDM 2024 test | TRE: 0.748 -> 0.725 | Improved |
| add DTMD diffusion losses (ablation) | PRN with centroid loss · ISICDM 2024 test | AAE: 0.670 -> 0.646 | Improved |
| increase centroid loss weight from 0.1 to 0.2 (ablation) | TAlignDiff · ISICDM 2024 test | TRE: 0.725 -> 0.766 | Hurt |
| increase denoising loss weight from 0.01 to 0.05 (ablation) | TAlignDiff · ISICDM 2024 test | TRE: 0.725 -> 0.739 | Hurt |
| decrease denoising loss weight from 0.01 to 0.005 (ablation) | TAlignDiff · ISICDM 2024 test | TRE: 0.725 -> 0.725 | No clear effect |
| joint training of PRN and DTMD instead of training separately first (ablation) | TAlignDiff · ISICDM 2024 | TRE and AAE | Improved |
| add DTMD module to baseline (ablation) | PRN baseline · ISICDM 2024 | TRE and AAE: 0.784 -> 0.725 | Improved |
| diffusion on transformation matrices instead of conditioning on raw point cloud features (TADPM) | tooth alignment · ISICDM 2024 test | TRE: 0.890 -> 0.725 | Improved |
Closest papers by meaning
- Collaborative Tooth Motion Diffusion Model in Digital Orthodontics (AAAI 2024 · similarity 0.86)
- Where Precision Meets Efficiency: Transformation Diffusion Model for Point Cloud Registration (AAAI 2025 · similarity 0.83)
- Point2SSM: Learning Morphological Variations of Anatomies from Point Clouds (ICLR 2024 · similarity 0.82)
- NeBLa: Neural Beer-Lambert for 3D Reconstruction of Oral Structures from Panoramic Radiographs (AAAI 2024 · similarity 0.82)
- SeaLion: Semantic Part-Aware Latent Point Diffusion Models for 3D Generation (CVPR 2025 · similarity 0.82)
- OralGPT-Plus: Learning to Use Visual Tools via Reinforcement Learning for Panoramic X-ray Analysis (CVPR 2026 · similarity 0.81)
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.
