3D Part Segmentation On Shapenet Part

评估指标

Class Average IoU

Instance Average IoU

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
GeomGCNN	-	89.1	Exploiting Local Geometry for Feature and Graph Construction for Better 3D Point Cloud Processing with Graph Neural Networks	-
Ours	-	88.1	Beyond local patches: Preserving global–local interactions by enhancing self-attention via 3D point cloud tokenization	-
AVS-Net	85.7	87.3	AVS-Net: Point Sampling with Adaptive Voxel Size for 3D Scene Understanding
SPoTr	85.4	87.2	Self-positioning Point-based Transformer for Point Cloud Understanding
Diffusion Unit	85.2	87.1	Diffusion Unit: Interpretable Edge Enhancement and Suppression Learning for 3D Point Cloud Segmentation
PointNeXt	85.2	87.1	PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies
CurveNet+GAM	-	87.0	$(0, 4)$ dualities	-
DeltaConv (U-ResNet)	-	86.9	DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds
PointMLP+TAP	85.2	86.9	Take-A-Photo: 3D-to-2D Generative Pre-training of Point Cloud Models
AGCN	85.7	86.9	AGCN: Adversarial Graph Convolutional Network for 3D Point Cloud Segmentation	-
PointVector-S(C=64)	-	86.9	PointVector: A Vector Representation In Point Cloud Analysis
CurveNet	-	86.8	Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis
Ps-CNN	83.4	86.8	Octree guided CNN with Spherical Kernels for 3D Point Clouds	-
OTMae3D	85.1	86.8	-	-
Spherical Kernel	84.9	86.8	Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds
MKConv	-	86.7	MKConv: Multidimensional Feature Representation for Point Cloud Analysis	-
PointGPT	84.8	86.6	-	-
PointTransformer	83.7	86.6	Point Transformer
Feature Geometric Net (FG-Net)	87.7	86.6	FG-Net: Fast Large-Scale LiDAR Point Clouds Understanding Network Leveraging Correlated Feature Mining and Geometric-Aware Modelling
DeltaNet	-	86.6	DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds

0 of 67 row(s) selected.