AI Paper Atlas

Research map / Point clouds / Point cloud classification

Transformer-based Tooth Alignment Prediction with Occlusion and Collision Constraints

Zhenxing Dong, Jiazhou Chen

ICCV 2025 · Point clouds › Point cloud classification · approach: Equivariant point cloud networks · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
add L_val to L_recon (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.64 -> 0.79Improved
add L_uni (occlusal distance uniformity) to L_recon + L_val (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.79 -> 0.81Improved
add L_fit (occlusal projecting overlap) to L_recon + L_val (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.79 -> 0.83Improved
add both occlusal losses L_uni and L_fit (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.79 -> 0.89Improved
add both occlusal losses (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)Mean rotation error: 8.3 -> 2.7Improved
use virtual arch line serialization instead of random order (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.77 -> 0.89Improved
use virtual arch line serialization instead of dental arch center (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.82 -> 0.89Improved
use Swin-T blocks (SWTBS) instead of Vision Transformer blocks (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.79 -> 0.90Improved
use Swin-T blocks instead of Swin-T v2 blocks (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.85 -> 0.90Improved
add SWTP multi-stage module (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.81 -> 0.90Improved
add tooth center info to our network (ablation)Ours · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.74 -> 0.90Improved
add tooth center info to baselines (ablation)TANet, PSTN, Ptv3 · Tooth alignment dataset (855 cases, 70 test)ADD/AUC: 0.82 -> 0.82No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy