The 3D semantic labeling task involves predicting a semantic labeling of a 3D scan mesh.

Evaluation and metrics

Our evaluation ranks all methods according to the PASCAL VOC intersection-over-union metric (IoU). IoU = TP/(TP+FP+FN), where TP, FP, and FN are the numbers of true positive, false positive, and false negative pixels, respectively. Predicted labels are evaluated per-vertex over the respective 3D scan mesh; for 3D approaches that operate on other representations like grids or points, the predicted labels should be mapped onto the mesh vertices (e.g., one such example for grid to mesh vertices is provided in the evaluation helpers).



This table lists the benchmark results for the 3D semantic label scenario.


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
Mix3Dpermissive0.781 10.964 10.855 10.843 120.781 30.858 90.575 40.831 240.685 90.714 20.979 10.594 40.310 200.801 10.892 110.841 20.819 30.723 30.940 90.887 20.725 17
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
Swin3Dpermissive0.779 20.861 160.818 110.836 150.790 10.875 20.576 30.905 30.704 30.739 10.969 70.611 10.349 60.756 150.958 10.702 340.805 120.708 60.916 240.898 10.801 1
PPT-SpUNet-Joint0.766 30.932 20.794 250.829 170.751 130.854 110.540 140.903 40.630 260.672 80.963 90.565 150.357 40.788 20.900 70.737 190.802 130.685 120.950 20.887 20.780 2
OctFormerpermissive0.766 30.925 30.808 170.849 60.786 20.846 170.566 60.876 110.690 70.674 70.960 110.576 110.226 560.753 170.904 50.777 60.815 60.722 40.923 200.877 80.776 4
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
CU-Hybrid Net0.764 50.924 40.819 90.840 130.757 90.853 120.580 10.848 180.709 20.643 160.958 140.587 80.295 260.753 170.884 150.758 130.815 60.725 20.927 190.867 140.743 9
OccuSeg+Semantic0.764 50.758 500.796 230.839 140.746 150.907 10.562 70.850 170.680 110.672 80.978 20.610 20.335 100.777 50.819 350.847 10.830 10.691 100.972 10.885 40.727 15
O-CNNpermissive0.762 70.924 40.823 60.844 110.770 50.852 130.577 20.847 200.711 10.640 200.958 140.592 50.217 620.762 120.888 120.758 130.813 80.726 10.932 170.868 130.744 8
Peng-Shuai Wang, Yang Liu, Yu-Xiao Guo, Chun-Yu Sun, Xin Tong: O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis. SIGGRAPH 2017
OA-CNN-L_ScanNet200.756 80.783 380.826 50.858 40.776 40.837 240.548 110.896 70.649 190.675 60.962 100.586 90.335 100.771 80.802 390.770 90.787 250.691 100.936 120.880 70.761 6
PointTransformerV20.752 90.742 570.809 160.872 10.758 80.860 80.552 90.891 80.610 340.687 30.960 110.559 180.304 230.766 100.926 30.767 100.797 170.644 260.942 70.876 110.722 19
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
DMF-Net0.752 90.906 80.793 270.802 330.689 300.825 350.556 80.867 130.681 100.602 340.960 110.555 200.365 30.779 40.859 200.747 160.795 210.717 50.917 230.856 220.764 5
PointConvFormer0.749 110.793 350.790 280.807 280.750 140.856 100.524 200.881 100.588 450.642 190.977 40.591 60.274 370.781 30.929 20.804 30.796 180.642 270.947 40.885 40.715 22
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
BPNetcopyleft0.749 110.909 60.818 110.811 250.752 110.839 230.485 370.842 210.673 120.644 150.957 170.528 290.305 220.773 70.859 200.788 40.818 50.693 90.916 240.856 220.723 18
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
MSP0.748 130.623 830.804 190.859 30.745 160.824 370.501 280.912 20.690 70.685 40.956 180.567 140.320 160.768 90.918 40.720 250.802 130.676 160.921 210.881 60.779 3
StratifiedFormerpermissive0.747 140.901 90.803 200.845 100.757 90.846 170.512 240.825 270.696 60.645 140.956 180.576 110.262 470.744 220.861 190.742 170.770 340.705 70.899 370.860 190.734 10
Xin Lai*, Jianhui Liu*, Li Jiang, Liwei Wang, Hengshuang Zhao, Shu Liu, Xiaojuan Qi, Jiaya Jia: Stratified Transformer for 3D Point Cloud Segmentation. CVPR 2022
VMNetpermissive0.746 150.870 140.838 20.858 40.729 210.850 150.501 280.874 120.587 460.658 120.956 180.564 160.299 240.765 110.900 70.716 280.812 90.631 320.939 100.858 200.709 23
Zeyu HU, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai: VMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation. ICCV 2021 (Oral)
Virtual MVFusion0.746 150.771 440.819 90.848 80.702 280.865 70.397 740.899 50.699 40.664 110.948 450.588 70.330 120.746 210.851 270.764 110.796 180.704 80.935 130.866 150.728 13
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
Retro-FPN0.744 170.842 220.800 210.767 450.740 170.836 260.541 130.914 10.672 130.626 230.958 140.552 210.272 390.777 50.886 140.696 350.801 150.674 180.941 80.858 200.717 20
EQ-Net0.743 180.620 840.799 220.849 60.730 200.822 390.493 350.897 60.664 140.681 50.955 210.562 170.378 10.760 130.903 60.738 180.801 150.673 190.907 290.877 80.745 7
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
SAT0.742 190.860 170.765 390.819 200.769 60.848 160.533 160.829 250.663 150.631 220.955 210.586 90.274 370.753 170.896 90.729 200.760 410.666 210.921 210.855 240.733 11
LRPNet0.742 190.816 300.806 180.807 280.752 110.828 330.575 40.839 230.699 40.637 210.954 260.520 310.320 160.755 160.834 310.760 120.772 310.676 160.915 260.862 170.717 20
TXC0.740 210.842 220.832 40.805 320.715 250.846 170.473 390.885 90.615 300.671 100.971 60.547 220.320 160.697 260.799 410.777 60.819 30.682 140.946 50.871 120.696 27
LargeKernel3D0.739 220.909 60.820 80.806 300.740 170.852 130.545 120.826 260.594 440.643 160.955 210.541 240.263 460.723 240.858 220.775 80.767 350.678 150.933 150.848 290.694 28
Yukang Chen*, Jianhui Liu*, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs. CVPR 2023
MinkowskiNetpermissive0.736 230.859 180.818 110.832 160.709 260.840 220.521 220.853 160.660 170.643 160.951 350.544 230.286 310.731 230.893 100.675 430.772 310.683 130.874 550.852 270.727 15
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
IPCA0.731 240.890 100.837 30.864 20.726 220.873 30.530 190.824 280.489 770.647 130.978 20.609 30.336 90.624 410.733 490.758 130.776 290.570 560.949 30.877 80.728 13
PointTransformer++0.725 250.727 650.811 150.819 200.765 70.841 210.502 270.814 330.621 290.623 250.955 210.556 190.284 320.620 420.866 170.781 50.757 440.648 240.932 170.862 170.709 23
SparseConvNet0.725 250.647 800.821 70.846 90.721 230.869 40.533 160.754 470.603 400.614 270.955 210.572 130.325 140.710 250.870 160.724 230.823 20.628 330.934 140.865 160.683 31
MatchingNet0.724 270.812 320.812 140.810 260.735 190.834 280.495 340.860 150.572 520.602 340.954 260.512 330.280 340.757 140.845 290.725 220.780 270.606 420.937 110.851 280.700 26
INS-Conv-semantic0.717 280.751 530.759 420.812 240.704 270.868 50.537 150.842 210.609 360.608 300.953 290.534 260.293 270.616 430.864 180.719 270.793 220.640 280.933 150.845 330.663 36
PointMetaBase0.714 290.835 240.785 290.821 180.684 320.846 170.531 180.865 140.614 310.596 380.953 290.500 360.246 520.674 270.888 120.692 360.764 370.624 340.849 700.844 340.675 33
contrastBoundarypermissive0.705 300.769 470.775 340.809 270.687 310.820 420.439 620.812 340.661 160.591 400.945 530.515 320.171 800.633 380.856 230.720 250.796 180.668 200.889 440.847 300.689 29
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
RFCR0.702 310.889 110.745 510.813 230.672 350.818 460.493 350.815 320.623 270.610 280.947 470.470 460.249 510.594 460.848 280.705 320.779 280.646 250.892 420.823 400.611 50
Jingyu Gong, Jiachen Xu, Xin Tan, Haichuan Song, Yanyun Qu, Yuan Xie, Lizhuang Ma: Omni-Supervised Point Cloud Segmentation via Gradual Receptive Field Component Reasoning. CVPR2021
One Thing One Click0.701 320.825 280.796 230.723 520.716 240.832 290.433 640.816 300.634 240.609 290.969 70.418 710.344 70.559 580.833 320.715 290.808 110.560 610.902 340.847 300.680 32
JSENetpermissive0.699 330.881 130.762 400.821 180.667 360.800 590.522 210.792 390.613 320.607 310.935 730.492 380.205 670.576 510.853 250.691 370.758 430.652 230.872 580.828 370.649 40
Zeyu HU, Mingmin Zhen, Xuyang BAI, Hongbo Fu, Chiew-lan Tai: JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds. ECCV 2020
One-Thing-One-Click0.693 340.743 560.794 250.655 760.684 320.822 390.497 330.719 570.622 280.617 260.977 40.447 580.339 80.750 200.664 650.703 330.790 240.596 460.946 50.855 240.647 41
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
PicassoNet-IIpermissive0.692 350.732 610.772 350.786 370.677 340.866 60.517 230.848 180.509 690.626 230.952 330.536 250.225 580.545 640.704 560.689 400.810 100.564 600.903 330.854 260.729 12
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
Feature_GeometricNetpermissive0.690 360.884 120.754 460.795 360.647 420.818 460.422 660.802 370.612 330.604 320.945 530.462 490.189 750.563 570.853 250.726 210.765 360.632 310.904 310.821 430.606 54
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
FusionNet0.688 370.704 700.741 550.754 490.656 380.829 310.501 280.741 520.609 360.548 470.950 390.522 300.371 20.633 380.756 440.715 290.771 330.623 350.861 660.814 450.658 37
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
Feature-Geometry Netpermissive0.685 380.866 150.748 480.819 200.645 440.794 620.450 510.802 370.587 460.604 320.945 530.464 480.201 700.554 600.840 300.723 240.732 530.602 440.907 290.822 420.603 57
KP-FCNN0.684 390.847 210.758 440.784 390.647 420.814 490.473 390.772 420.605 380.594 390.935 730.450 560.181 780.587 470.805 380.690 380.785 260.614 380.882 480.819 440.632 46
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
DGNet0.684 390.712 690.784 300.782 410.658 370.835 270.499 320.823 290.641 210.597 370.950 390.487 390.281 330.575 520.619 680.647 560.764 370.620 370.871 610.846 320.688 30
VACNN++0.684 390.728 640.757 450.776 420.690 290.804 560.464 450.816 300.577 510.587 410.945 530.508 350.276 360.671 280.710 540.663 480.750 470.589 510.881 490.832 360.653 39
Superpoint Network0.683 420.851 200.728 590.800 350.653 400.806 540.468 420.804 350.572 520.602 340.946 500.453 550.239 550.519 690.822 330.689 400.762 400.595 480.895 400.827 380.630 47
PointContrast_LA_SEM0.683 420.757 510.784 300.786 370.639 460.824 370.408 690.775 410.604 390.541 490.934 770.532 270.269 420.552 610.777 420.645 590.793 220.640 280.913 270.824 390.671 34
VI-PointConv0.676 440.770 460.754 460.783 400.621 500.814 490.552 90.758 450.571 540.557 450.954 260.529 280.268 440.530 670.682 600.675 430.719 560.603 430.888 450.833 350.665 35
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
ROSMRF3D0.673 450.789 360.748 480.763 470.635 480.814 490.407 710.747 490.581 500.573 420.950 390.484 400.271 410.607 440.754 450.649 530.774 300.596 460.883 470.823 400.606 54
SALANet0.670 460.816 300.770 370.768 440.652 410.807 530.451 480.747 490.659 180.545 480.924 830.473 450.149 900.571 540.811 370.635 620.746 480.623 350.892 420.794 570.570 67
PointConvpermissive0.666 470.781 390.759 420.699 610.644 450.822 390.475 380.779 400.564 570.504 650.953 290.428 650.203 690.586 490.754 450.661 490.753 450.588 520.902 340.813 470.642 42
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointASNLpermissive0.666 470.703 710.781 320.751 510.655 390.830 300.471 410.769 430.474 800.537 510.951 350.475 440.279 350.635 360.698 590.675 430.751 460.553 660.816 770.806 490.703 25
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PPCNN++permissive0.663 490.746 540.708 630.722 530.638 470.820 420.451 480.566 840.599 420.541 490.950 390.510 340.313 190.648 330.819 350.616 670.682 720.590 500.869 620.810 480.656 38
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
DCM-Net0.658 500.778 400.702 660.806 300.619 510.813 520.468 420.693 650.494 720.524 570.941 640.449 570.298 250.510 710.821 340.675 430.727 550.568 580.826 750.803 510.637 44
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
HPGCNN0.656 510.698 720.743 530.650 770.564 680.820 420.505 260.758 450.631 250.479 700.945 530.480 420.226 560.572 530.774 430.690 380.735 510.614 380.853 690.776 720.597 60
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
SAFNet-segpermissive0.654 520.752 520.734 570.664 740.583 630.815 480.399 730.754 470.639 220.535 530.942 620.470 460.309 210.665 290.539 740.650 520.708 610.635 300.857 680.793 590.642 42
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Netpermissive0.645 530.778 400.731 580.699 610.577 640.829 310.446 530.736 530.477 790.523 590.945 530.454 530.269 420.484 780.749 480.618 650.738 490.599 450.827 740.792 620.621 49
MVPNetpermissive0.641 540.831 250.715 610.671 710.590 590.781 680.394 750.679 670.642 200.553 460.937 700.462 490.256 480.649 320.406 870.626 630.691 690.666 210.877 510.792 620.608 53
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
PointConv-SFPN0.641 540.776 420.703 650.721 540.557 710.826 340.451 480.672 690.563 580.483 690.943 610.425 680.162 850.644 340.726 500.659 500.709 600.572 550.875 530.786 670.559 72
PointMRNet0.640 560.717 680.701 670.692 640.576 650.801 580.467 440.716 580.563 580.459 750.953 290.429 640.169 820.581 500.854 240.605 680.710 580.550 670.894 410.793 590.575 65
FPConvpermissive0.639 570.785 370.760 410.713 590.603 540.798 600.392 760.534 890.603 400.524 570.948 450.457 510.250 500.538 650.723 520.598 720.696 670.614 380.872 580.799 520.567 69
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
PD-Net0.638 580.797 340.769 380.641 820.590 590.820 420.461 460.537 880.637 230.536 520.947 470.388 790.206 660.656 300.668 630.647 560.732 530.585 530.868 630.793 590.473 91
PointSPNet0.637 590.734 600.692 740.714 580.576 650.797 610.446 530.743 510.598 430.437 800.942 620.403 750.150 890.626 400.800 400.649 530.697 660.557 640.846 710.777 710.563 70
SConv0.636 600.830 260.697 700.752 500.572 670.780 700.445 550.716 580.529 630.530 540.951 350.446 590.170 810.507 730.666 640.636 610.682 720.541 720.886 460.799 520.594 61
Supervoxel-CNN0.635 610.656 780.711 620.719 550.613 520.757 790.444 580.765 440.534 620.566 430.928 810.478 430.272 390.636 350.531 760.664 470.645 820.508 800.864 650.792 620.611 50
joint point-basedpermissive0.634 620.614 850.778 330.667 730.633 490.825 350.420 670.804 350.467 820.561 440.951 350.494 370.291 280.566 550.458 820.579 790.764 370.559 630.838 720.814 450.598 59
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
PointMTL0.632 630.731 620.688 770.675 680.591 580.784 670.444 580.565 850.610 340.492 670.949 430.456 520.254 490.587 470.706 550.599 710.665 780.612 410.868 630.791 660.579 64
APCF-Net0.631 640.742 570.687 790.672 690.557 710.792 650.408 690.665 700.545 600.508 620.952 330.428 650.186 760.634 370.702 570.620 640.706 620.555 650.873 560.798 540.581 63
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
3DSM_DMMF0.631 640.626 820.745 510.801 340.607 530.751 800.506 250.729 560.565 560.491 680.866 970.434 600.197 730.595 450.630 670.709 310.705 630.560 610.875 530.740 820.491 86
PointNet2-SFPN0.631 640.771 440.692 740.672 690.524 760.837 240.440 610.706 630.538 610.446 770.944 590.421 700.219 610.552 610.751 470.591 750.737 500.543 710.901 360.768 740.557 73
FusionAwareConv0.630 670.604 870.741 550.766 460.590 590.747 810.501 280.734 540.503 710.527 550.919 870.454 530.323 150.550 630.420 860.678 420.688 700.544 690.896 390.795 560.627 48
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DenSeR0.628 680.800 330.625 890.719 550.545 740.806 540.445 550.597 790.448 860.519 600.938 690.481 410.328 130.489 770.499 810.657 510.759 420.592 490.881 490.797 550.634 45
SegGroup_sempermissive0.627 690.818 290.747 500.701 600.602 550.764 760.385 800.629 760.490 750.508 620.931 800.409 730.201 700.564 560.725 510.618 650.692 680.539 730.873 560.794 570.548 76
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
SIConv0.625 700.830 260.694 720.757 480.563 690.772 740.448 520.647 730.520 650.509 610.949 430.431 630.191 740.496 750.614 690.647 560.672 760.535 750.876 520.783 680.571 66
HPEIN0.618 710.729 630.668 800.647 790.597 570.766 750.414 680.680 660.520 650.525 560.946 500.432 610.215 630.493 760.599 700.638 600.617 870.570 560.897 380.806 490.605 56
Li Jiang, Hengshuang Zhao, Shu Liu, Xiaoyong Shen, Chi-Wing Fu, Jiaya Jia: Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation. ICCV 2019
SPH3D-GCNpermissive0.610 720.858 190.772 350.489 940.532 750.792 650.404 720.643 750.570 550.507 640.935 730.414 720.046 990.510 710.702 570.602 700.705 630.549 680.859 670.773 730.534 79
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
AttAN0.609 730.760 490.667 810.649 780.521 770.793 630.457 470.648 720.528 640.434 820.947 470.401 760.153 880.454 800.721 530.648 550.717 570.536 740.904 310.765 750.485 87
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
wsss-transformer0.600 740.634 810.743 530.697 630.601 560.781 680.437 630.585 820.493 730.446 770.933 780.394 770.011 1010.654 310.661 660.603 690.733 520.526 760.832 730.761 770.480 88
dtc_net0.596 750.683 730.725 600.715 570.549 730.803 570.444 580.647 730.493 730.495 660.941 640.409 730.000 1030.424 850.544 730.598 720.703 650.522 770.912 280.792 620.520 82
LAP-D0.594 760.720 660.692 740.637 830.456 860.773 730.391 780.730 550.587 460.445 790.940 670.381 800.288 290.434 830.453 840.591 750.649 800.581 540.777 810.749 810.610 52
DPC0.592 770.720 660.700 680.602 870.480 820.762 780.380 810.713 610.585 490.437 800.940 670.369 820.288 290.434 830.509 800.590 770.639 850.567 590.772 820.755 790.592 62
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
CCRFNet0.589 780.766 480.659 840.683 660.470 850.740 830.387 790.620 780.490 750.476 710.922 850.355 850.245 530.511 700.511 790.571 800.643 830.493 840.872 580.762 760.600 58
ROSMRF0.580 790.772 430.707 640.681 670.563 690.764 760.362 830.515 900.465 830.465 740.936 720.427 670.207 650.438 810.577 710.536 830.675 750.486 850.723 880.779 690.524 81
SD-DETR0.576 800.746 540.609 930.445 980.517 780.643 940.366 820.714 600.456 840.468 730.870 960.432 610.264 450.558 590.674 610.586 780.688 700.482 860.739 860.733 840.537 78
SQN_0.1%0.569 810.676 750.696 710.657 750.497 790.779 710.424 650.548 860.515 670.376 870.902 940.422 690.357 40.379 880.456 830.596 740.659 790.544 690.685 910.665 950.556 74
TextureNetpermissive0.566 820.672 770.664 820.671 710.494 800.719 840.445 550.678 680.411 920.396 850.935 730.356 840.225 580.412 860.535 750.565 810.636 860.464 880.794 800.680 920.568 68
Jingwei Huang, Haotian Zhang, Li Yi, Thomas Funkerhouser, Matthias Niessner, Leonidas Guibas: TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes. CVPR
DVVNet0.562 830.648 790.700 680.770 430.586 620.687 880.333 870.650 710.514 680.475 720.906 910.359 830.223 600.340 900.442 850.422 940.668 770.501 810.708 890.779 690.534 79
Pointnet++ & Featurepermissive0.557 840.735 590.661 830.686 650.491 810.744 820.392 760.539 870.451 850.375 880.946 500.376 810.205 670.403 870.356 900.553 820.643 830.497 820.824 760.756 780.515 83
GMLPs0.538 850.495 950.693 730.647 790.471 840.793 630.300 900.477 910.505 700.358 890.903 930.327 880.081 960.472 790.529 770.448 920.710 580.509 780.746 840.737 830.554 75
PanopticFusion-label0.529 860.491 960.688 770.604 860.386 910.632 950.225 1000.705 640.434 890.293 950.815 980.348 860.241 540.499 740.669 620.507 850.649 800.442 940.796 790.602 980.561 71
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
subcloud_weak0.516 870.676 750.591 960.609 840.442 870.774 720.335 860.597 790.422 910.357 900.932 790.341 870.094 950.298 920.528 780.473 900.676 740.495 830.602 970.721 870.349 98
Online SegFusion0.515 880.607 860.644 870.579 890.434 880.630 960.353 840.628 770.440 870.410 830.762 1010.307 900.167 830.520 680.403 880.516 840.565 900.447 920.678 920.701 890.514 84
Davide Menini, Suryansh Kumar, Martin R. Oswald, Erik Sandstroem, Cristian Sminchisescu, Luc van Gool: A Real-Time Learning Framework for Joint 3D Reconstruction and Semantic Segmentation. Robotics and Automation Letters Submission
3DMV, FTSDF0.501 890.558 910.608 940.424 1000.478 830.690 870.246 960.586 810.468 810.450 760.911 890.394 770.160 860.438 810.212 970.432 930.541 950.475 870.742 850.727 850.477 89
PCNN0.498 900.559 900.644 870.560 910.420 900.711 860.229 980.414 920.436 880.352 910.941 640.324 890.155 870.238 970.387 890.493 860.529 960.509 780.813 780.751 800.504 85
3DMV0.484 910.484 970.538 980.643 810.424 890.606 990.310 880.574 830.433 900.378 860.796 990.301 910.214 640.537 660.208 980.472 910.507 990.413 970.693 900.602 980.539 77
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PointCNN with RGBpermissive0.458 920.577 890.611 920.356 1020.321 990.715 850.299 920.376 960.328 990.319 930.944 590.285 930.164 840.216 1000.229 950.484 880.545 940.456 900.755 830.709 880.475 90
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
FCPNpermissive0.447 930.679 740.604 950.578 900.380 920.682 890.291 930.106 1020.483 780.258 1000.920 860.258 970.025 1000.231 990.325 910.480 890.560 920.463 890.725 870.666 940.231 102
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
DGCNN_reproducecopyleft0.446 940.474 980.623 900.463 960.366 940.651 920.310 880.389 950.349 970.330 920.937 700.271 950.126 920.285 930.224 960.350 990.577 890.445 930.625 950.723 860.394 94
Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon: Dynamic Graph CNN for Learning on Point Clouds. TOG 2019
PNET20.442 950.548 920.548 970.597 880.363 950.628 970.300 900.292 970.374 940.307 940.881 950.268 960.186 760.238 970.204 990.407 950.506 1000.449 910.667 930.620 970.462 92
SurfaceConvPF0.442 950.505 940.622 910.380 1010.342 970.654 910.227 990.397 940.367 950.276 970.924 830.240 980.198 720.359 890.262 930.366 960.581 880.435 950.640 940.668 930.398 93
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
Tangent Convolutionspermissive0.438 970.437 1000.646 860.474 950.369 930.645 930.353 840.258 990.282 1010.279 960.918 880.298 920.147 910.283 940.294 920.487 870.562 910.427 960.619 960.633 960.352 97
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
3DWSSS0.425 980.525 930.647 850.522 920.324 980.488 1020.077 1030.712 620.353 960.401 840.636 1030.281 940.176 790.340 900.565 720.175 1030.551 930.398 980.370 1030.602 980.361 96
SPLAT Netcopyleft0.393 990.472 990.511 990.606 850.311 1000.656 900.245 970.405 930.328 990.197 1010.927 820.227 1000.000 1030.001 1040.249 940.271 1020.510 970.383 1000.593 980.699 900.267 100
Hang Su, Varun Jampani, Deqing Sun, Subhransu Maji, Evangelos Kalogerakis, Ming-Hsuan Yang, Jan Kautz: SPLATNet: Sparse Lattice Networks for Point Cloud Processing. CVPR 2018
ScanNet+FTSDF0.383 1000.297 1020.491 1000.432 990.358 960.612 980.274 940.116 1010.411 920.265 980.904 920.229 990.079 970.250 950.185 1000.320 1000.510 970.385 990.548 990.597 1010.394 94
PointNet++permissive0.339 1010.584 880.478 1010.458 970.256 1020.360 1030.250 950.247 1000.278 1020.261 990.677 1020.183 1010.117 930.212 1010.145 1020.364 970.346 1030.232 1030.548 990.523 1020.252 101
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNetpermissive0.308 1020.353 1010.290 1030.278 1030.166 1030.553 1000.169 1020.286 980.147 1030.148 1030.908 900.182 1020.064 980.023 1030.018 1040.354 980.363 1010.345 1010.546 1010.685 910.278 99
ScanNetpermissive0.306 1030.203 1030.366 1020.501 930.311 1000.524 1010.211 1010.002 1040.342 980.189 1020.786 1000.145 1030.102 940.245 960.152 1010.318 1010.348 1020.300 1020.460 1020.437 1030.182 103
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
ERROR0.054 1040.000 1040.041 1040.172 1040.030 1040.062 1040.001 1040.035 1030.004 1040.051 1040.143 1040.019 1040.003 1020.041 1020.050 1030.003 1040.054 1040.018 1040.005 1040.264 1040.082 104