ML Research Wiki / Benchmarks / 3D Semantic Segmentation / SensatUrban

SensatUrban

3D Semantic Segmentation Benchmark

Performance Over Time

📊 Showing 7 results | 📏 Metric: mIoU

Top Performing Models

Rank Model Paper mIoU Date Code
1 LCPFormer LCPFormer: Towards Effective 3D Point Cloud Analysis via Local Context Propagation in Transformers 63.40 2022-10-23 📦 zhh6425/LocalContextPropagation
2 EyeNet Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scene 62.30 2023-01-30 📦 Yacovitch/EyeNet
3 BEV-Seg3D-Net Efficient Urban-scale Point Clouds Segmentation with BEV Projection 61.70 2021-09-19 📦 zouzhenhong98/SensatUrban-BEV-Seg3D
4 KPConv KPConv: Flexible and Deformable Convolution for Point Clouds 57.58 2019-04-18 📦 isl-org/Open3D-ML 📦 HuguesTHOMAS/KPConv-PyTorch 📦 HuguesTHOMAS/KPConv
5 SparseConv 3D Semantic Segmentation with Submanifold Sparse Convolutional Networks 42.66 2017-11-28 📦 facebookresearch/SparseConvNet 📦 edraizen/molmimic 📦 btgraham/SparseConvNet
6 SPGraph Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs 37.29 2017-11-27 📦 loicland/superpoint_graph 📦 jsgaobiao/superpoint_graph
7 TangentConv Tangent Convolutions for Dense Prediction in 3D 33.30 2018-07-06 📦 tatarchm/tangent_conv

All Papers (7)