3D Object Classification On Modelnet40

Classification Accuracy

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

		Paper Title
Ours	93.6	Exploiting Inductive Bias in Transformer for Point Cloud Classification and Segmentation
G3DNet-18 MLP, Fine-Tuned, Vote	91.7	General-Purpose Deep Point Cloud Feature Extractor
CrossMoCo	91.49	CrossMoCo: Multi-modal Momentum Contrastive Learning for Point Cloud
O-CNN(6)	89.9	O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis
Spherical Kernel	89.3	Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds
3D-PointCapsNet	89.3	3D Point Capsule Networks
ECC (12 votes)	83.2	Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs

0 of 7 row(s) selected.