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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysort bysorted 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 80.781 10.858 60.575 20.831 110.685 40.714 10.979 10.594 30.310 120.801 10.892 40.841 20.819 30.723 20.940 40.887 10.725 8
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
O-CNNpermissive0.762 30.924 20.823 40.844 70.770 20.852 70.577 10.847 70.711 10.640 100.958 60.592 40.217 430.762 50.888 50.758 50.813 50.726 10.932 100.868 50.744 2
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
StratifiedFormer0.747 50.901 40.803 100.845 60.757 30.846 90.512 100.825 120.696 30.645 70.956 80.576 60.262 300.744 100.861 80.742 70.770 220.705 30.899 210.860 80.734 3
BPNetcopyleft0.749 40.909 30.818 70.811 140.752 40.839 110.485 200.842 90.673 60.644 80.957 70.528 150.305 140.773 30.859 90.788 30.818 40.693 50.916 110.856 100.723 9
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
OccuSeg+Semantic0.764 20.758 350.796 120.839 90.746 50.907 10.562 30.850 60.680 50.672 30.978 20.610 10.335 60.777 20.819 210.847 10.830 10.691 60.972 10.885 20.727 6
MatchingNet0.724 120.812 210.812 90.810 150.735 60.834 140.495 170.860 40.572 320.602 200.954 130.512 170.280 220.757 70.845 170.725 120.780 160.606 260.937 60.851 130.700 12
EQ-Net0.743 80.620 610.799 110.849 30.730 70.822 220.493 180.897 20.664 70.681 20.955 110.562 90.378 10.760 60.903 10.738 80.801 90.673 80.907 140.877 30.745 1
VMNetpermissive0.746 60.870 100.838 20.858 20.729 80.850 80.501 130.874 30.587 270.658 50.956 80.564 80.299 150.765 40.900 20.716 150.812 60.631 170.939 50.858 90.709 10
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)
IPCA0.731 100.890 50.837 30.864 10.726 90.873 20.530 70.824 130.489 550.647 60.978 20.609 20.336 50.624 220.733 330.758 50.776 180.570 380.949 20.877 30.728 4
SparseConvNet0.725 110.647 580.821 50.846 50.721 100.869 30.533 60.754 300.603 240.614 130.955 110.572 70.325 90.710 120.870 60.724 130.823 20.628 180.934 80.865 70.683 14
One Thing One Click0.701 150.825 190.796 120.723 360.716 110.832 150.433 420.816 140.634 120.609 150.969 50.418 520.344 30.559 390.833 180.715 160.808 70.560 420.902 180.847 150.680 15
MinkowskiNetpermissive0.736 90.859 120.818 70.832 100.709 120.840 100.521 90.853 50.660 80.643 90.951 190.544 100.286 210.731 110.893 30.675 270.772 200.683 70.874 400.852 120.727 6
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
PicassoNet-IIpermissive0.696 170.704 490.790 150.787 220.709 120.837 120.459 290.815 160.543 420.615 120.956 80.529 130.250 330.551 420.790 250.703 200.799 100.619 210.908 130.848 140.700 12
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
INS-Conv-semantic0.717 130.751 380.759 240.812 130.704 140.868 40.537 50.842 90.609 200.608 160.953 150.534 110.293 170.616 230.864 70.719 140.793 120.640 120.933 90.845 160.663 18
Virtual MVFusion0.746 60.771 300.819 60.848 40.702 150.865 50.397 530.899 10.699 20.664 40.948 280.588 50.330 70.746 90.851 140.764 40.796 110.704 40.935 70.866 60.728 4
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
VACNN++0.684 230.728 450.757 270.776 260.690 160.804 380.464 280.816 140.577 310.587 240.945 360.508 180.276 240.671 130.710 370.663 320.750 310.589 340.881 340.832 190.653 20
One-Thing-One-Click0.693 180.743 390.794 140.655 560.684 170.822 220.497 160.719 400.622 150.617 110.977 40.447 410.339 40.750 80.664 450.703 200.790 140.596 290.946 30.855 110.647 23
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
CU-Hybrid Net0.693 180.596 650.789 160.803 170.677 180.800 400.469 240.846 80.554 400.591 230.948 280.500 190.316 110.609 240.847 160.732 90.808 70.593 320.894 250.839 170.652 21
RFCR0.702 140.889 60.745 320.813 120.672 190.818 260.493 180.815 160.623 140.610 140.947 310.470 280.249 350.594 270.848 150.705 190.779 170.646 110.892 270.823 230.611 33
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
JSENetpermissive0.699 160.881 90.762 220.821 110.667 200.800 400.522 80.792 220.613 160.607 170.935 540.492 210.205 470.576 330.853 110.691 220.758 280.652 100.872 420.828 200.649 22
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
FusionNet0.688 220.704 490.741 350.754 330.656 210.829 170.501 130.741 350.609 200.548 300.950 230.522 160.371 20.633 200.756 280.715 160.771 210.623 190.861 470.814 280.658 19
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
PointASNLpermissive0.666 300.703 510.781 180.751 350.655 220.830 160.471 230.769 260.474 580.537 330.951 190.475 260.279 230.635 180.698 410.675 270.751 300.553 470.816 570.806 310.703 11
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
Superpoint Network0.683 250.851 140.728 400.800 190.653 230.806 360.468 250.804 180.572 320.602 200.946 330.453 380.239 380.519 480.822 190.689 250.762 260.595 310.895 240.827 210.630 29
SALANet0.670 290.816 200.770 210.768 290.652 240.807 350.451 310.747 320.659 90.545 310.924 620.473 270.149 680.571 350.811 220.635 440.746 320.623 190.892 270.794 400.570 50
KP-FCNN0.684 230.847 150.758 260.784 240.647 250.814 300.473 220.772 250.605 220.594 220.935 540.450 390.181 570.587 280.805 230.690 230.785 150.614 220.882 320.819 270.632 28
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
Feature-Geometry Netpermissive0.690 200.884 70.754 280.795 200.647 250.818 260.422 430.802 200.612 170.604 180.945 360.462 310.189 530.563 370.853 110.726 100.765 230.632 150.904 150.821 250.606 37
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
Feature_GeometricNetpermissive0.690 200.884 70.754 280.795 200.647 250.818 260.422 430.802 200.612 170.604 180.945 360.462 310.189 530.563 370.853 110.726 100.765 230.632 150.904 150.821 250.606 37
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
PointConvpermissive0.666 300.781 250.759 240.699 420.644 280.822 220.475 210.779 230.564 370.504 450.953 150.428 470.203 490.586 300.754 290.661 330.753 290.588 350.902 180.813 300.642 24
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointContrast_LA_SEM0.683 250.757 360.784 170.786 230.639 290.824 210.408 480.775 240.604 230.541 320.934 580.532 120.269 270.552 400.777 260.645 410.793 120.640 120.913 120.824 220.671 16
ROSMRF3D0.673 280.789 230.748 310.763 310.635 300.814 300.407 500.747 320.581 300.573 250.950 230.484 220.271 260.607 250.754 290.649 370.774 190.596 290.883 310.823 230.606 37
joint point-basedpermissive0.634 430.614 620.778 190.667 530.633 310.825 200.420 450.804 180.467 600.561 270.951 190.494 200.291 180.566 360.458 600.579 560.764 250.559 440.838 520.814 280.598 42
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
VI-PointConv0.676 270.770 320.754 280.783 250.621 320.814 300.552 40.758 280.571 340.557 280.954 130.529 130.268 290.530 460.682 420.675 270.719 370.603 270.888 290.833 180.665 17
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
DCM-Net0.658 320.778 260.702 450.806 160.619 330.813 330.468 250.693 470.494 530.524 380.941 460.449 400.298 160.510 500.821 200.675 270.727 360.568 400.826 540.803 330.637 26
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
Supervoxel-CNN0.635 420.656 560.711 420.719 380.613 340.757 580.444 390.765 270.534 440.566 260.928 600.478 250.272 250.636 170.531 540.664 310.645 600.508 580.864 460.792 430.611 33
3DSM_DMMF0.631 460.626 600.745 320.801 180.607 350.751 590.506 110.729 390.565 360.491 470.866 740.434 430.197 510.595 260.630 470.709 180.705 450.560 420.875 380.740 610.491 65
FPConvpermissive0.639 390.785 240.760 230.713 410.603 360.798 430.392 550.534 670.603 240.524 380.948 280.457 340.250 330.538 440.723 350.598 520.696 480.614 220.872 420.799 340.567 52
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
HPEIN0.618 520.729 440.668 580.647 590.597 370.766 550.414 460.680 490.520 470.525 370.946 330.432 440.215 440.493 550.599 490.638 420.617 650.570 380.897 220.806 310.605 40
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
PointMTL0.632 450.731 430.688 550.675 480.591 380.784 490.444 390.565 650.610 190.492 460.949 250.456 350.254 320.587 280.706 380.599 510.665 570.612 250.868 450.791 460.579 47
MVPNetpermissive0.641 360.831 160.715 410.671 510.590 390.781 500.394 540.679 500.642 100.553 290.937 520.462 310.256 310.649 150.406 650.626 450.691 490.666 90.877 360.792 430.608 36
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
FusionAwareConv0.630 490.604 640.741 350.766 300.590 390.747 600.501 130.734 370.503 510.527 360.919 660.454 360.323 100.550 430.420 640.678 260.688 500.544 500.896 230.795 380.627 30
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DVVNet0.562 600.648 570.700 470.770 280.586 410.687 670.333 650.650 540.514 490.475 520.906 700.359 610.223 410.340 680.442 630.422 720.668 560.501 590.708 680.779 490.534 59
SAFNet-segpermissive0.654 340.752 370.734 370.664 540.583 420.815 290.399 520.754 300.639 110.535 340.942 440.470 280.309 130.665 140.539 520.650 360.708 430.635 140.857 490.793 410.642 24
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Netpermissive0.645 350.778 260.731 380.699 420.577 430.829 170.446 340.736 360.477 570.523 400.945 360.454 360.269 270.484 570.749 320.618 480.738 330.599 280.827 530.792 430.621 31
PointSPNet0.637 400.734 420.692 520.714 400.576 440.797 440.446 340.743 340.598 260.437 580.942 440.403 540.150 670.626 210.800 240.649 370.697 470.557 450.846 510.777 510.563 53
MCCNNpermissive0.633 440.866 110.731 380.771 270.576 440.809 340.410 470.684 480.497 520.491 470.949 250.466 300.105 720.581 310.646 460.620 460.680 520.542 520.817 560.795 380.618 32
P. Hermosilla, T. Ritschel, P.P. Vazquez, A. Vinacua, T. Ropinski: Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds. SIGGRAPH Asia 2018
PointMRNet0.640 380.717 480.701 460.692 440.576 440.801 390.467 270.716 410.563 380.459 540.953 150.429 460.169 600.581 310.854 100.605 490.710 400.550 480.894 250.793 410.575 48
SConv0.636 410.830 170.697 490.752 340.572 470.780 510.445 360.716 410.529 450.530 350.951 190.446 420.170 590.507 520.666 440.636 430.682 510.541 530.886 300.799 340.594 44
HPGCNN0.656 330.698 520.743 340.650 570.564 480.820 250.505 120.758 280.631 130.479 500.945 360.480 240.226 390.572 340.774 270.690 230.735 350.614 220.853 500.776 520.597 43
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
ROSMRF0.580 580.772 290.707 430.681 470.563 490.764 560.362 610.515 680.465 610.465 530.936 530.427 490.207 460.438 600.577 500.536 610.675 540.486 630.723 670.779 490.524 61
SIConv0.625 510.830 170.694 500.757 320.563 490.772 540.448 330.647 560.520 470.509 420.949 250.431 450.191 520.496 540.614 480.647 400.672 550.535 550.876 370.783 480.571 49
PointConv-SFPN0.641 360.776 280.703 440.721 370.557 510.826 190.451 310.672 520.563 380.483 490.943 430.425 500.162 630.644 160.726 340.659 340.709 420.572 370.875 380.786 470.559 55
APCF-Net0.631 460.742 400.687 570.672 490.557 510.792 470.408 480.665 530.545 410.508 430.952 180.428 470.186 550.634 190.702 390.620 460.706 440.555 460.873 410.798 360.581 46
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
DenSeR0.628 500.800 220.625 680.719 380.545 530.806 360.445 360.597 600.448 640.519 410.938 510.481 230.328 80.489 560.499 590.657 350.759 270.592 330.881 340.797 370.634 27
SPH3D-GCNpermissive0.610 530.858 130.772 200.489 730.532 540.792 470.404 510.643 570.570 350.507 440.935 540.414 530.046 780.510 500.702 390.602 500.705 450.549 490.859 480.773 530.534 59
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
PointNet2-SFPN0.631 460.771 300.692 520.672 490.524 550.837 120.440 410.706 450.538 430.446 560.944 410.421 510.219 420.552 400.751 310.591 530.737 340.543 510.901 200.768 540.557 56
AttAN0.609 540.760 340.667 590.649 580.521 560.793 450.457 300.648 550.528 460.434 600.947 310.401 550.153 660.454 590.721 360.648 390.717 380.536 540.904 150.765 550.485 66
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
TextureNetpermissive0.566 590.672 550.664 600.671 510.494 570.719 630.445 360.678 510.411 700.396 630.935 540.356 620.225 400.412 640.535 530.565 580.636 640.464 660.794 600.680 700.568 51
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
Pointnet++ & Featurepermissive0.557 610.735 410.661 610.686 450.491 580.744 610.392 550.539 660.451 630.375 650.946 330.376 580.205 470.403 650.356 680.553 600.643 610.497 600.824 550.756 570.515 62
DPC0.592 560.720 460.700 470.602 660.480 590.762 570.380 600.713 430.585 290.437 580.940 480.369 590.288 190.434 620.509 580.590 550.639 630.567 410.772 620.755 580.592 45
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
3DMV, FTSDF0.501 670.558 690.608 710.424 770.478 600.690 660.246 730.586 630.468 590.450 550.911 680.394 560.160 640.438 600.212 740.432 710.541 720.475 650.742 650.727 630.477 67
GMLPs0.538 630.495 730.693 510.647 590.471 610.793 450.300 670.477 690.505 500.358 670.903 720.327 660.081 750.472 580.529 550.448 700.710 400.509 560.746 640.737 620.554 57
CCRFNet0.589 570.766 330.659 620.683 460.470 620.740 620.387 590.620 590.490 540.476 510.922 640.355 630.245 360.511 490.511 570.571 570.643 610.493 620.872 420.762 560.600 41
LAP-D0.594 550.720 460.692 520.637 620.456 630.773 530.391 570.730 380.587 270.445 570.940 480.381 570.288 190.434 620.453 610.591 530.649 580.581 360.777 610.749 600.610 35
subcloud_weak0.516 650.676 540.591 730.609 630.442 640.774 520.335 640.597 600.422 690.357 680.932 590.341 650.094 740.298 700.528 560.473 680.676 530.495 610.602 740.721 640.349 75
Online SegFusion0.515 660.607 630.644 660.579 680.434 650.630 730.353 620.628 580.440 650.410 610.762 780.307 680.167 610.520 470.403 660.516 620.565 670.447 700.678 700.701 670.514 63
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
PointMRNet-lite0.553 620.633 590.648 630.659 550.430 660.800 400.390 580.592 620.454 620.371 660.939 500.368 600.136 700.368 660.448 620.560 590.715 390.486 630.882 320.720 650.462 69
3DMV0.484 690.484 750.538 750.643 610.424 670.606 760.310 660.574 640.433 680.378 640.796 760.301 690.214 450.537 450.208 750.472 690.507 760.413 740.693 690.602 750.539 58
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PCNN0.498 680.559 680.644 660.560 700.420 680.711 650.229 750.414 700.436 660.352 690.941 460.324 670.155 650.238 740.387 670.493 640.529 730.509 560.813 580.751 590.504 64
PanopticFusion-label0.529 640.491 740.688 550.604 650.386 690.632 720.225 770.705 460.434 670.293 720.815 750.348 640.241 370.499 530.669 430.507 630.649 580.442 710.796 590.602 750.561 54
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
FCPNpermissive0.447 710.679 530.604 720.578 690.380 700.682 680.291 700.106 790.483 560.258 770.920 650.258 740.025 790.231 760.325 690.480 670.560 690.463 670.725 660.666 720.231 79
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
Tangent Convolutionspermissive0.438 740.437 770.646 650.474 740.369 710.645 710.353 620.258 760.282 780.279 730.918 670.298 700.147 690.283 710.294 700.487 650.562 680.427 730.619 730.633 730.352 74
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
PNET20.442 720.548 700.548 740.597 670.363 720.628 740.300 670.292 740.374 720.307 710.881 730.268 730.186 550.238 740.204 760.407 730.506 770.449 690.667 710.620 740.462 69
ScanNet+FTSDF0.383 770.297 790.491 770.432 760.358 730.612 750.274 710.116 780.411 700.265 750.904 710.229 760.079 760.250 720.185 770.320 770.510 740.385 760.548 760.597 780.394 72
SurfaceConvPF0.442 720.505 720.622 690.380 780.342 740.654 700.227 760.397 720.367 730.276 740.924 620.240 750.198 500.359 670.262 710.366 740.581 660.435 720.640 720.668 710.398 71
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
3DWSSS0.425 750.525 710.647 640.522 710.324 750.488 790.077 800.712 440.353 740.401 620.636 800.281 720.176 580.340 680.565 510.175 800.551 700.398 750.370 800.602 750.361 73
PointCNN with RGBpermissive0.458 700.577 670.611 700.356 790.321 760.715 640.299 690.376 730.328 760.319 700.944 410.285 710.164 620.216 770.229 730.484 660.545 710.456 680.755 630.709 660.475 68
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
SPLAT Netcopyleft0.393 760.472 760.511 760.606 640.311 770.656 690.245 740.405 710.328 760.197 780.927 610.227 770.000 810.001 810.249 720.271 790.510 740.383 770.593 750.699 680.267 77
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
ScanNetpermissive0.306 800.203 800.366 790.501 720.311 770.524 780.211 780.002 810.342 750.189 790.786 770.145 800.102 730.245 730.152 780.318 780.348 790.300 790.460 790.437 800.182 80
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
PointNet++permissive0.339 780.584 660.478 780.458 750.256 790.360 800.250 720.247 770.278 790.261 760.677 790.183 780.117 710.212 780.145 790.364 750.346 800.232 800.548 760.523 790.252 78
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNetpermissive0.308 790.353 780.290 800.278 800.166 800.553 770.169 790.286 750.147 800.148 800.908 690.182 790.064 770.023 800.018 810.354 760.363 780.345 780.546 780.685 690.278 76
ERROR0.054 810.000 810.041 810.172 810.030 810.062 810.001 810.035 800.004 810.051 810.143 810.019 810.003 800.041 790.050 800.003 810.054 810.018 810.005 810.264 810.082 81

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




Method Infoavg ap 50%bathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sort bysort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
SoftGroup0.761 11.000 10.808 90.845 40.716 10.862 30.243 60.824 20.655 30.620 10.734 10.699 20.791 20.981 150.716 30.844 10.769 11.000 10.594 3
IPCA-Inst0.731 21.000 10.788 130.884 30.698 20.788 150.252 50.760 80.646 40.511 60.637 20.665 30.804 11.000 10.644 90.778 50.747 21.000 10.561 6
HAISpermissive0.699 31.000 10.849 30.820 50.675 30.808 90.279 40.757 90.465 90.517 50.596 30.559 50.600 131.000 10.654 80.767 60.676 60.994 170.560 7
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
SphereSeg0.680 51.000 10.856 20.744 110.618 40.893 10.151 90.651 140.713 10.537 40.579 70.430 140.651 41.000 10.389 240.744 130.697 30.991 180.601 2
Mask-Group0.664 71.000 10.822 60.764 100.616 50.815 60.139 130.694 120.597 50.459 110.566 80.599 40.600 130.516 290.715 40.819 30.635 101.000 10.603 1
DENet0.629 161.000 10.797 110.608 170.589 60.627 230.219 70.882 10.310 200.402 220.383 250.396 180.650 51.000 10.663 60.543 290.691 51.000 10.568 5
INS-Conv-instance0.657 81.000 10.760 170.667 140.581 70.863 20.323 30.655 130.477 70.473 90.549 100.432 130.650 51.000 10.655 70.738 140.585 150.944 220.472 15
RWSeg0.567 200.528 310.708 240.626 150.580 80.745 190.063 220.627 150.240 230.400 230.497 150.464 90.515 201.000 10.475 180.745 120.571 161.000 10.429 19
OccuSeg+instance0.672 61.000 10.758 190.682 130.576 90.842 40.477 10.504 250.524 60.567 20.585 60.451 100.557 181.000 10.751 20.797 40.563 181.000 10.467 16
DD-UNet+Group0.635 150.667 250.797 120.714 120.562 100.774 170.146 100.810 50.429 120.476 80.546 120.399 170.633 121.000 10.632 100.722 160.609 121.000 10.514 8
H. Liu, R. Liu, K. Yang, J. Zhang, K. Peng, R. Stiefelhagen: HIDA: Towards Holistic Indoor Understanding for the Visually Impaired via Semantic Instance Segmentation with a Wearable Solid-State LiDAR Sensor. ICCVW 2021
SSTNetpermissive0.698 41.000 10.697 250.888 20.556 110.803 100.387 20.626 160.417 130.556 30.585 50.702 10.600 131.000 10.824 10.720 170.692 41.000 10.509 10
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
RPGN0.643 111.000 10.758 180.582 220.539 120.826 50.046 240.765 70.372 160.436 160.588 40.539 70.650 51.000 10.577 120.750 110.653 90.997 140.495 13
PE0.645 101.000 10.773 140.798 70.538 130.786 160.088 200.799 60.350 180.435 170.547 110.545 60.646 110.933 160.562 140.761 90.556 220.997 140.501 12
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
Dyco3Dcopyleft0.641 121.000 10.841 40.893 10.531 140.802 110.115 170.588 210.448 100.438 140.537 130.430 150.550 190.857 180.534 150.764 80.657 70.987 190.568 4
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
3D-MPA0.611 171.000 10.833 50.765 90.526 150.756 180.136 150.588 210.470 80.438 150.432 210.358 210.650 50.857 180.429 200.765 70.557 201.000 10.430 18
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
CSC-Pretrained0.648 91.000 10.810 70.768 80.523 160.813 70.143 120.819 30.389 140.422 180.511 140.443 110.650 51.000 10.624 110.732 150.634 111.000 10.375 22
Occipital-SCS0.512 231.000 10.716 210.509 230.506 170.611 240.092 190.602 200.177 260.346 250.383 240.165 260.442 240.850 220.386 250.618 250.543 230.889 260.389 21
PointGroup0.636 141.000 10.765 150.624 160.505 180.797 120.116 160.696 110.384 150.441 130.559 90.476 80.596 161.000 10.666 50.756 100.556 210.997 140.513 9
Li Jiang, Hengshuang Zhao, Shaoshuai Shi, Shu Liu, Chi-Wing Fu, Jiaya Jia: PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation. CVPR 2020 [oral]
GICN0.638 131.000 10.895 10.800 60.480 190.676 200.144 110.737 100.354 170.447 120.400 230.365 200.700 31.000 10.569 130.836 20.599 131.000 10.473 14
SSEN0.575 191.000 10.761 160.473 240.477 200.795 130.066 210.529 230.658 20.460 100.461 180.380 190.331 280.859 170.401 230.692 210.653 81.000 10.348 24
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
Sparse R-CNN0.515 221.000 10.538 310.282 270.468 210.790 140.173 80.345 270.429 110.413 210.484 160.176 250.595 170.591 270.522 160.668 230.476 250.986 200.327 25
PCJC0.578 181.000 10.810 80.583 210.449 220.813 80.042 250.603 190.341 190.490 70.465 170.410 160.650 50.835 230.264 280.694 200.561 190.889 260.504 11
MASCpermissive0.447 270.528 310.555 290.381 250.382 230.633 220.002 310.509 240.260 220.361 240.432 200.327 220.451 220.571 280.367 260.639 240.386 260.980 210.276 27
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
R-PointNet0.306 310.500 330.405 340.311 260.348 240.589 250.054 230.068 340.126 280.283 280.290 270.028 340.219 310.214 320.331 270.396 340.275 310.821 310.245 28
MTML0.549 211.000 10.807 100.588 200.327 250.647 210.004 300.815 40.180 250.418 190.364 260.182 240.445 231.000 10.442 190.688 220.571 171.000 10.396 20
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
3D-BoNet0.488 241.000 10.672 260.590 190.301 260.484 320.098 180.620 170.306 210.341 260.259 280.125 280.434 250.796 240.402 220.499 310.513 240.909 250.439 17
Bo Yang, Jianan Wang, Ronald Clark, Qingyong Hu, Sen Wang, Andrew Markham, Niki Trigoni: Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds. NeurIPS 2019 Spotlight
SALoss-ResNet0.459 261.000 10.737 200.159 340.259 270.587 260.138 140.475 260.217 240.416 200.408 220.128 270.315 290.714 250.411 210.536 300.590 140.873 290.304 26
Zhidong Liang, Ming Yang, Hao Li, Chunxiang Wang: 3D Instance Embedding Learning With a Structure-Aware Loss Function for Point Cloud Segmentation. IEEE Robotics and Automation Letters (IROS2020)
PanopticFusion-inst0.478 250.667 250.712 230.595 180.259 280.550 290.000 330.613 180.175 270.250 300.434 190.437 120.411 270.857 180.485 170.591 280.267 330.944 220.359 23
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
3D-SISpermissive0.382 281.000 10.432 330.245 290.190 290.577 270.013 280.263 290.033 330.320 270.240 290.075 300.422 260.857 180.117 310.699 180.271 320.883 280.235 29
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
UNet-backbone0.319 300.667 250.715 220.233 300.189 300.479 330.008 290.218 300.067 320.201 310.173 310.107 290.123 330.438 300.150 290.615 260.355 270.916 240.093 35
RandSA0.250 320.333 340.613 270.229 310.163 310.493 300.000 330.304 280.107 290.147 330.100 320.052 330.231 300.119 330.039 330.445 330.325 280.654 320.141 32
Hier3Dcopyleft0.323 290.667 250.542 300.264 280.157 320.550 280.000 330.205 320.009 340.270 290.218 300.075 300.500 210.688 260.007 360.698 190.301 300.459 340.200 30
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
Region0.248 330.667 250.437 320.188 320.153 330.491 310.000 330.208 310.094 310.153 320.099 330.057 320.217 320.119 330.039 330.466 320.302 290.640 330.140 33
Sgpn_scannet0.143 350.208 360.390 350.169 330.065 340.275 350.029 260.069 330.000 350.087 350.043 340.014 360.027 360.000 350.112 320.351 350.168 350.438 350.138 34
MaskRCNN 2d->3d Proj0.058 360.333 340.002 360.000 360.053 350.002 360.002 320.021 360.000 350.045 360.024 360.238 230.065 350.000 350.014 350.107 360.020 360.110 360.006 36
3D-BEVIS0.248 330.667 250.566 280.076 350.035 360.394 340.027 270.035 350.098 300.099 340.030 350.025 350.098 340.375 310.126 300.604 270.181 340.854 300.171 31
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.

This table lists the benchmark results for the 2D semantic label scenario.


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
Virtual MVFusion (R)0.745 10.861 10.839 10.881 10.672 10.512 10.422 100.898 10.723 10.714 10.954 20.454 10.509 10.773 10.895 10.756 10.820 10.653 10.935 10.891 10.728 1
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
BPNet_2Dcopyleft0.670 20.822 30.795 30.836 20.659 20.481 20.451 60.769 20.656 30.567 30.931 30.395 30.390 40.700 20.534 30.689 50.770 20.574 30.865 30.831 30.675 2
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia and Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
CU-Hybrid-2D Net0.636 30.825 20.820 20.179 160.648 30.463 30.549 10.742 30.676 20.628 20.961 10.420 20.379 50.684 30.381 100.732 20.723 30.599 20.827 80.851 20.634 3
DMMF_3d0.605 40.651 60.744 70.782 30.637 40.387 40.536 20.732 40.590 50.540 40.856 130.359 60.306 100.596 60.539 20.627 120.706 40.497 60.785 120.757 110.476 13
MCA-Net0.595 60.533 120.756 50.746 50.590 50.334 80.506 30.670 70.587 60.500 80.905 80.366 50.352 60.601 50.506 60.669 100.648 60.501 50.839 70.769 90.516 12
DMMF0.597 50.543 110.755 60.749 40.585 60.338 60.494 40.704 60.598 40.494 100.911 60.347 80.327 90.593 70.527 40.675 70.646 80.513 40.842 60.774 80.527 11
RFBNet0.592 70.616 70.758 40.659 60.581 70.330 90.469 50.655 100.543 90.524 50.924 40.355 70.336 80.572 80.479 80.671 80.648 60.480 70.814 100.814 40.614 6
SSMAcopyleft0.577 90.695 40.716 100.439 120.563 80.314 100.444 80.719 50.551 70.503 70.887 100.346 90.348 70.603 40.353 120.709 30.600 110.457 100.901 20.786 60.599 7
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019
SN_RN152pyrx8_RVCcopyleft0.546 100.572 90.663 130.638 80.518 90.298 120.366 150.633 120.510 110.446 120.864 110.296 110.267 120.542 100.346 130.704 40.575 140.431 120.853 50.766 100.630 4
FuseNetpermissive0.535 110.570 100.681 120.182 150.512 100.290 130.431 90.659 90.504 120.495 90.903 90.308 100.428 30.523 120.365 110.676 60.621 100.470 90.762 130.779 70.541 9
Caner Hazirbas, Lingni Ma, Csaba Domokos, Daniel Cremers: FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-based CNN Architecture. ACCV 2016
DCRedNet0.583 80.682 50.723 80.542 100.510 110.310 110.451 60.668 80.549 80.520 60.920 50.375 40.446 20.528 110.417 90.670 90.577 130.478 80.862 40.806 50.628 5
ILC-PSPNet0.475 150.490 140.581 150.289 140.507 120.067 170.379 130.610 140.417 150.435 130.822 160.278 120.267 120.503 130.228 140.616 140.533 150.375 140.820 90.729 120.560 8
3DMV (2d proj)0.498 130.481 150.612 140.579 90.456 130.343 50.384 120.623 130.525 100.381 140.845 140.254 130.264 140.557 90.182 150.581 150.598 120.429 130.760 140.661 160.446 15
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
MSeg1080_RVCpermissive0.485 140.505 130.709 110.092 170.427 140.241 140.411 110.654 110.385 170.457 110.861 120.053 170.279 110.503 130.481 70.645 110.626 90.365 150.748 150.725 130.529 10
John Lambert*, Zhuang Liu*, Ozan Sener, James Hays, Vladlen Koltun: MSeg: A Composite Dataset for Multi-domain Semantic Segmentation. CVPR 2020
Enet (reimpl)0.376 160.264 170.452 170.452 110.365 150.181 150.143 170.456 160.409 160.346 160.769 170.164 150.218 150.359 160.123 170.403 170.381 170.313 170.571 160.685 150.472 14
Re-implementation of Adam Paszke, Abhishek Chaurasia, Sangpil Kim, Eugenio Culurciello: ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation.
ScanNet (2d proj)permissive0.330 170.293 160.521 160.657 70.361 160.161 160.250 160.004 170.440 140.183 170.836 150.125 160.060 170.319 170.132 160.417 160.412 160.344 160.541 170.427 170.109 17
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
AdapNet++copyleft0.503 120.613 80.722 90.418 130.358 170.337 70.370 140.479 150.443 130.368 150.907 70.207 140.213 160.464 150.525 50.618 130.657 50.450 110.788 110.721 140.408 16
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019

This table lists the benchmark results for the 2D semantic instance scenario.




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sort bysort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
UniDet_RVC0.205 10.381 10.323 10.037 10.226 10.177 10.063 10.277 10.120 10.067 10.131 10.074 20.317 10.080 10.235 10.289 10.141 10.678 10.080 1
MaskRCNN_ScanNetpermissive0.119 20.129 20.212 20.002 20.112 20.148 20.014 20.205 20.044 20.066 20.078 20.095 10.142 20.030 20.128 20.139 20.080 20.459 20.057 2
Re-implementation of Kaiming He, Georgia Gkioxari, Piotr Dollár, Ross Girshick: Mask R-CNN. ICCV'17

This table lists the benchmark results for the scene type classification scenario.




Method Infoavg recallapartmentbathroombedroom / hotelbookstore / libraryconference roomcopy/mail roomhallwaykitchenlaundry roomliving room / loungemiscofficestorage / basement / garage
sort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
SE-ResNeXt-SSMA0.498 30.000 40.812 30.941 10.500 20.500 30.500 20.500 20.429 40.500 20.667 20.500 10.625 30.000 2
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. arXiv
multi-taskpermissive0.700 10.500 11.000 10.882 20.500 21.000 11.000 10.500 21.000 11.000 10.778 10.000 20.938 10.000 2
Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler: Indoor Scene Recognition in 3D. IROS 2020
3DASPP-SCE0.691 20.500 10.938 20.824 31.000 11.000 10.500 21.000 10.857 20.500 20.556 30.000 20.812 20.500 1
resnet50_scannet0.353 40.250 30.812 30.529 40.500 20.500 30.000 40.500 20.571 30.000 40.556 30.000 20.375 40.000 2