AI Paper Atlas

Research map / Point clouds / Registration and matching

Neural Intrinsic Embedding for Non-Rigid Point Cloud Matching

Puhua Jiang, Mingze Sun, Ruqi Huang

CVPR 2023 · Point clouds › Registration and matching · approach: Non-rigid point set registration · 16 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 the bijectivity loss to the relative geodesic loss (ablation)NIE · SCAPE_r embedding evaluationOPT: 4.4 -> 3.5Improved
Add the KL loss to the relative geodesic and bijectivity losses (ablation)NIE · SCAPE_r embedding evaluationOPT: 3.5 -> 3.3Improved
Use modified DGCNN sampling aggregation (ablation)NIE · SCAPE_r embedding evaluationOPT: 3.3 -> 3.1Improved
Add the bijectivity loss (ablation)NIE · SCAPE_r embedding evaluationrelative geodesic error: 8.8 -> 12.4Hurt
Add the KL loss after the relative geodesic and bijectivity losses (ablation)NIE · SCAPE_r embedding evaluationrelative geodesic error: 12.4 -> 10.6Improved
Add bijectivity and KL losses to the relative geodesic loss (ablation)NIE · final matching on SCAPE_rmatching error: 13.2 -> 11.5Improved
Increase basis dimension from 10 to 20 (ablation)NIE · basis dimension ablationOPT: 4.7 -> 3.1Improved
Increase basis dimension from 20 to 30 (ablation)NIE · basis dimension ablationrelative geodesic error: 9.5 -> 11.3Hurt
Train with relative rather than absolute geodesic errorNIE embedding · SCAPE_r test shapesrelative geodesic error: 12.0 -> 9.5Improved
Train on SURREAL_r and test on re-meshed FAUST_rNIE + NIM · FAUST_r matchingmean geodesic error: 15. -> 6.5Improved
Add Gaussian noise to test point cloudsNIE + NIM · FAUST_r matching trained on SURREAL_rmean geodesic error: 6.5 -> 7.2Hurt
Match partial point clouds with holesNIE + NIM · partial point cloud matching: holemean geodesic error: 15. -> 7.0Improved

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