Semantic Segmentation on ScanNet (test)
78.5mIoUPonderV2
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PonderV2#Params.=41.0M2023.10 | 78.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Swin3D-LInput=Mesh Sampled Point Cloud2024.01 | 77.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OctformerInput=Mesh Sampled Point Cloud2024.01 | 76.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet + PPT#Params.=41.0M2023.10 | 76.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MVNetBackbone=SR-UNet2025.12 | 75.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Point Transformer v2Input=Mesh Sampled Point Cloud2024.01 | 75.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2#Params.=12.8M2023.10 | 75.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BPNetInput=Sensor RGBD Point Cloud2024.01 | 74.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Stratified TransformerInput=Mesh Sampled Point Cloud2024.01 | 74.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VMVFInput=Rendered RGBD Point Cloud2024.01 | 74.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ODIN-Swin-BInput=Sensor RGBD Point Cloud2024.01 | 74.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepViewAggInput=Sensor RGBD Point Cloud2024.01 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointContrastBackbone=SR-UNet2025.12 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SceneContextBackbone=SR-UNet2025.12 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| StratifiedFormer#Params.=18.8M2023.10 | 73.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MinkowskiNet2020.07 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet#Params.=39.2M2023.10 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MM-3DSceneBackbone=PointTrans2025.12 | 72.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sparse Conv Net2020.07 | 72.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointNext#Params.=41.6M2023.10 | 71.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DepthContrastBackbone=SR-UNet2025.12 | 71.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| JSENet2020.07 | 69.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| KPConvType=rigid2020.07 | 68.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| KPConvType=deform2020.07 | 68.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointConv2020.07 | 66.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MVPNet2020.07 | 64.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MVPNetInput=Sensor RGBD Point Cloud2024.01 | 64.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MCCNN2020.07 | 63.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HPEIN2020.07 | 61.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SPH3D-GCN2020.07 | 61 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RFBNetMultimodal=true, Backbone=AdapNet++2019.06 | 59.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSMAMultimodal=true, Backbone=AdapNet++2019.06 | 57.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSMAMultimodal=true2018.08 | 57.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TextureNet2020.07 | 56.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointConv2018.11 | 55.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Panoptic Fusion2020.07 | 52.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FuseNetMultimodal=true2019.06 | 52.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FuseNetMultimodal=true2018.08 | 52.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AdapNet++Multimodal=true2019.06 | 50.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AdapNet++Multimodal=false2018.08 | 50.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3DMV (2d proj)Multimodal=true2019.06 | 49.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3DMV (2d proj)Multimodal=true2018.08 | 49.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangSurfOpen-vocabulary=true2024.12 | 47.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 66.89 | 33.25 | 51.4 | 40.09 | |
| PSPNetMultimodal=false2019.06 | 47.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PSPNetMultimodal=false2018.08 | 47.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointCNN2020.07 | 45.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FCPN2020.07 | 44.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tangent Convolutions2018.11 | 43.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TangentConv2020.07 | 43.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tangent ConvolutionsInput signals=Depth (D)2018.07 | 40.9 | 80.9 | 52.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tangent ConvolutionsInput signals=Depth (D) + Height (H) + Normals (N) + Color (RGB)2018.07 | 40.9 | 80.1 | 55.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tangent ConvolutionsInput signals=Depth (D) + Height (H) + Normals (N)2018.07 | 40.7 | 80.3 | 55.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tangent ConvolutionsInput signals=Depth (D) + Height (H)2018.07 | 40.3 | 80.6 | 52.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SPLAT Netvariant=SPLATNet3D2018.11 | 39.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EnetMultimodal=false2018.08 | 37.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATSegOpen-vocabulary=true2024.12 | 36.52 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 42.36 | 33.66 | 45.57 | 24.51 | |
| PointNet++2018.11 | 33.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ScanNet2018.11 | 30.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GS-GroupOpen-vocabulary=true2024.12 | 26.86 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 43.41 | 20.78 | 23.84 | 19.42 | |
| LSegOpen-vocabulary=true2024.12 | 25.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 32.75 | 19.11 | 33.83 | 15.53 | |
| LangSplatOpen-vocabulary=true2024.12 | 22.32 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 36.69 | 16.81 | 24.29 | 11.49 | |
| OctNet2018.07 | 18.1 | 76.6 | 26.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ScanNet2018.07 | 13.5 | 69.4 | 19.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointNet2018.07 | 12.2 | 68.1 | 17.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3DMVnum_views=5, training=end-to-end2018.03 | — | 75 | — | 73.9 | 95.6 | 69.9 | 80.7 | 85.9 | 75.8 | 67.8 | 86.6 | 61.2 | 88.1 | 55.8 | 31.9 | 73.2 | 82.4 | 74.8 | 82.6 | 88.3 | 72.8 | 94.7 | 58.5 | — | — | — | — | |
| 4l-spec-cpDomain=Local k-NN Graphs, Kernel=graph conv2018.03 | — | 84.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Dai et al.Domain=Voxel Grid, Kernel=3D conv2018.03 | — | 73 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointNetDomain=Full Points, Kernel=point-MLP2018.03 | — | 73.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Pointnet++Domain=Local Points, Kernel=point-MLP2018.03 | — | 84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointNet++2018.03 | — | 60.2 | — | 89.5 | 97.8 | 39.8 | 69.7 | 86 | 68.3 | 59.6 | 27.5 | 23.7 | 84.3 | 0 | 37.6 | 66.7 | 48.7 | 54.7 | 85 | 84.8 | 62.8 | 86.1 | 30.7 | — | — | — | — | |
| ScanComplete2018.03 | — | 52.8 | — | 87.2 | 96.9 | 44.5 | 65.7 | 75.1 | 72.1 | 63.8 | 13.6 | 16.9 | 70.5 | 10.4 | 31.4 | 40.9 | 49.8 | 38.7 | 46.8 | 72.2 | 47.4 | 85.1 | 26.9 | — | — | — | — | |
| ScanNet2018.03 | — | 50.8 | — | 70.1 | 90.3 | 49.8 | 62.4 | 69.3 | 75.7 | 68.4 | 48.9 | 20.1 | 64.6 | 3.4 | 32.1 | 36.8 | 7 | 66.4 | 46.8 | 69.9 | 39.4 | 74.3 | 19.5 | — | — | — | — |