Semantic Segmentation On S3Dis Area5

评估指标

Number of params
mAcc
mIoU
oAcc

评测结果

各个模型在此基准测试上的表现结果

模型名称
Number of params
mAcc
mIoU
oAcc
Paper TitleRepository
SPoTrN/A76.470.890.7Self-positioning Point-based Transformer for Point Cloud Understanding-
PointVector-XL-78.172.391PointVector: A Vector Representation In Point Cloud Analysis-
DITR--74.1-DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation-
KPConv14.1M72.867.1-KPConv: Flexible and Deformable Convolution for Point Clouds-
PointMixer6.5M77.471.4-PointMixer: MLP-Mixer for Point Cloud Understanding-
TangentConvN/A62.2--Tangent Convolutions for Dense Prediction in 3D-
SSP+SPG290K68.261.787.9Point Cloud Oversegmentation with Graph-Structured Deep Metric Learning-
DPCN/A-61.28-Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds-
HPEINN/A68.361.8587.18Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation-
SegCloudN/A57.448.9-SEGCloud: Semantic Segmentation of 3D Point Clouds-
WindowNorm+PointTransformerN/A77.971.491.1Window Normalization: Enhancing Point Cloud Understanding by Unifying Inconsistent Point Densities-
SuperCluster0.21-68.1-Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering-
Swin3D-LN/A80.574.592.7Swin3D: A Pretrained Transformer Backbone for 3D Indoor Scene Understanding-
PointNetN/A-41.1-Point Transformer-
SPG(PTv2)-79.573.391.9Subspace Prototype Guidance for Mitigating Class Imbalance in Point Cloud Semantic Segmentation-
Pamba--73.5-Pamba: Enhancing Global Interaction in Point Clouds via State Space Model-
ConDaFormer-78.973.592.4ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding-
Serialized Piont Mamba--70.6-Serialized Point Mamba: A Serialized Point Cloud Mamba Segmentation Model-
SCF-NetN/A71.863.787.2SCF-Net: Learning Spatial Contextual Features for Large-Scale Point Cloud Segmentation
Superpoint Transformer212K77.368.989.5Efficient 3D Semantic Segmentation with Superpoint Transformer-
0 of 60 row(s) selected.