HyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAI超神経

3D Dense Shape Correspondence On Shrec 19

評価指標

Accuracy at 1%
Euclidean Mean Error (EME)

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Accuracy at 1%
Euclidean Mean Error (EME)
Paper TitleRepository
CorrNet3D (Trained on Surreal)6.06.9CorrNet3D: Unsupervised End-to-end Learning of Dense Correspondence for 3D Point Clouds
CorrNet3D0.433.8CorrNet3D: Unsupervised End-to-end Learning of Dense Correspondence for 3D Point Clouds
Elementery Structures(Trained on Surreal)2.37.6Learning elementary structures for 3D shape generation and matching
DPC15.35.6DPC: Unsupervised Deep Point Correspondence via Cross and Self Construction
SE-ORNet17.55.1SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence
Diff-FMaps (Trained on Surreal)4.07.1Correspondence Learning via Linearly-invariant Embedding
TANet (Trained on Surreal)21.54.5Unsupervised Template-assisted Point Cloud Shape Correspondence Network-
DPC (Trained on Surreal)17.76.1DPC: Unsupervised Deep Point Correspondence via Cross and Self Construction
3DCODED (Trained on Surreal)2.18.13D-CODED : 3D Correspondences by Deep Deformation
SE-ORNet (Trained on Surreal)21.54.6SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence
Diffusion 3D Features (Zero-shot)26.41.7Diffusion 3D Features (Diff3F): Decorating Untextured Shapes with Distilled Semantic Features
0 of 11 row(s) selected.