Benchmark Results - ScanNet Benchmark

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

Method	avg iou	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	floor	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	wall	window

Mix3D	0.781 1	0.964 1	0.855 1	0.843 8	0.781 1	0.858 6	0.575 2	0.831 11	0.685 4	0.714 1	0.979 1	0.594 3	0.310 12	0.801 1	0.892 4	0.841 2	0.819 3	0.723 2	0.940 4	0.887 1	0.725 8
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
O-CNN	0.762 3	0.924 2	0.823 4	0.844 7	0.770 2	0.852 7	0.577 1	0.847 7	0.711 1	0.640 10	0.958 6	0.592 4	0.217 43	0.762 5	0.888 5	0.758 5	0.813 5	0.726 1	0.932 10	0.868 5	0.744 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
StratifiedFormer	0.747 5	0.901 4	0.803 10	0.845 6	0.757 3	0.846 9	0.512 10	0.825 12	0.696 3	0.645 7	0.956 8	0.576 6	0.262 30	0.744 10	0.861 8	0.742 7	0.770 22	0.705 3	0.899 21	0.860 8	0.734 3

BPNet	0.749 4	0.909 3	0.818 7	0.811 14	0.752 4	0.839 11	0.485 20	0.842 9	0.673 6	0.644 8	0.957 7	0.528 15	0.305 14	0.773 3	0.859 9	0.788 3	0.818 4	0.693 5	0.916 11	0.856 10	0.723 9
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
OccuSeg+Semantic	0.764 2	0.758 35	0.796 12	0.839 9	0.746 5	0.907 1	0.562 3	0.850 6	0.680 5	0.672 3	0.978 2	0.610 1	0.335 6	0.777 2	0.819 21	0.847 1	0.830 1	0.691 6	0.972 1	0.885 2	0.727 6

MatchingNet	0.724 12	0.812 21	0.812 9	0.810 15	0.735 6	0.834 14	0.495 17	0.860 4	0.572 32	0.602 20	0.954 13	0.512 17	0.280 22	0.757 7	0.845 17	0.725 12	0.780 16	0.606 26	0.937 6	0.851 13	0.700 12

EQ-Net	0.743 8	0.620 61	0.799 11	0.849 3	0.730 7	0.822 22	0.493 18	0.897 2	0.664 7	0.681 2	0.955 11	0.562 9	0.378 1	0.760 6	0.903 1	0.738 8	0.801 9	0.673 8	0.907 14	0.877 3	0.745 1

VMNet	0.746 6	0.870 10	0.838 2	0.858 2	0.729 8	0.850 8	0.501 13	0.874 3	0.587 27	0.658 5	0.956 8	0.564 8	0.299 15	0.765 4	0.900 2	0.716 15	0.812 6	0.631 17	0.939 5	0.858 9	0.709 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)
IPCA	0.731 10	0.890 5	0.837 3	0.864 1	0.726 9	0.873 2	0.530 7	0.824 13	0.489 55	0.647 6	0.978 2	0.609 2	0.336 5	0.624 22	0.733 33	0.758 5	0.776 18	0.570 38	0.949 2	0.877 3	0.728 4

SparseConvNet	0.725 11	0.647 58	0.821 5	0.846 5	0.721 10	0.869 3	0.533 6	0.754 30	0.603 24	0.614 13	0.955 11	0.572 7	0.325 9	0.710 12	0.870 6	0.724 13	0.823 2	0.628 18	0.934 8	0.865 7	0.683 14

One Thing One Click	0.701 15	0.825 19	0.796 12	0.723 36	0.716 11	0.832 15	0.433 42	0.816 14	0.634 12	0.609 15	0.969 5	0.418 52	0.344 3	0.559 39	0.833 18	0.715 16	0.808 7	0.560 42	0.902 18	0.847 15	0.680 15

MinkowskiNet	0.736 9	0.859 12	0.818 7	0.832 10	0.709 12	0.840 10	0.521 9	0.853 5	0.660 8	0.643 9	0.951 19	0.544 10	0.286 21	0.731 11	0.893 3	0.675 27	0.772 20	0.683 7	0.874 40	0.852 12	0.727 6
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
PicassoNet-II	0.696 17	0.704 49	0.790 15	0.787 22	0.709 12	0.837 12	0.459 29	0.815 16	0.543 42	0.615 12	0.956 8	0.529 13	0.250 33	0.551 42	0.790 25	0.703 20	0.799 10	0.619 21	0.908 13	0.848 14	0.700 12
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
INS-Conv-semantic	0.717 13	0.751 38	0.759 24	0.812 13	0.704 14	0.868 4	0.537 5	0.842 9	0.609 20	0.608 16	0.953 15	0.534 11	0.293 17	0.616 23	0.864 7	0.719 14	0.793 12	0.640 12	0.933 9	0.845 16	0.663 18

Virtual MVFusion	0.746 6	0.771 30	0.819 6	0.848 4	0.702 15	0.865 5	0.397 53	0.899 1	0.699 2	0.664 4	0.948 28	0.588 5	0.330 7	0.746 9	0.851 14	0.764 4	0.796 11	0.704 4	0.935 7	0.866 6	0.728 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 23	0.728 45	0.757 27	0.776 26	0.690 16	0.804 38	0.464 28	0.816 14	0.577 31	0.587 24	0.945 36	0.508 18	0.276 24	0.671 13	0.710 37	0.663 32	0.750 31	0.589 34	0.881 34	0.832 19	0.653 20

One-Thing-One-Click	0.693 18	0.743 39	0.794 14	0.655 56	0.684 17	0.822 22	0.497 16	0.719 40	0.622 15	0.617 11	0.977 4	0.447 41	0.339 4	0.750 8	0.664 45	0.703 20	0.790 14	0.596 29	0.946 3	0.855 11	0.647 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 Net	0.693 18	0.596 65	0.789 16	0.803 17	0.677 18	0.800 40	0.469 24	0.846 8	0.554 40	0.591 23	0.948 28	0.500 19	0.316 11	0.609 24	0.847 16	0.732 9	0.808 7	0.593 32	0.894 25	0.839 17	0.652 21

RFCR	0.702 14	0.889 6	0.745 32	0.813 12	0.672 19	0.818 26	0.493 18	0.815 16	0.623 14	0.610 14	0.947 31	0.470 28	0.249 35	0.594 27	0.848 15	0.705 19	0.779 17	0.646 11	0.892 27	0.823 23	0.611 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
JSENet	0.699 16	0.881 9	0.762 22	0.821 11	0.667 20	0.800 40	0.522 8	0.792 22	0.613 16	0.607 17	0.935 54	0.492 21	0.205 47	0.576 33	0.853 11	0.691 22	0.758 28	0.652 10	0.872 42	0.828 20	0.649 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
FusionNet	0.688 22	0.704 49	0.741 35	0.754 33	0.656 21	0.829 17	0.501 13	0.741 35	0.609 20	0.548 30	0.950 23	0.522 16	0.371 2	0.633 20	0.756 28	0.715 16	0.771 21	0.623 19	0.861 47	0.814 28	0.658 19
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
PointASNL	0.666 30	0.703 51	0.781 18	0.751 35	0.655 22	0.830 16	0.471 23	0.769 26	0.474 58	0.537 33	0.951 19	0.475 26	0.279 23	0.635 18	0.698 41	0.675 27	0.751 30	0.553 47	0.816 57	0.806 31	0.703 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 Network	0.683 25	0.851 14	0.728 40	0.800 19	0.653 23	0.806 36	0.468 25	0.804 18	0.572 32	0.602 20	0.946 33	0.453 38	0.239 38	0.519 48	0.822 19	0.689 25	0.762 26	0.595 31	0.895 24	0.827 21	0.630 29

SALANet	0.670 29	0.816 20	0.770 21	0.768 29	0.652 24	0.807 35	0.451 31	0.747 32	0.659 9	0.545 31	0.924 62	0.473 27	0.149 68	0.571 35	0.811 22	0.635 44	0.746 32	0.623 19	0.892 27	0.794 40	0.570 50

KP-FCNN	0.684 23	0.847 15	0.758 26	0.784 24	0.647 25	0.814 30	0.473 22	0.772 25	0.605 22	0.594 22	0.935 54	0.450 39	0.181 57	0.587 28	0.805 23	0.690 23	0.785 15	0.614 22	0.882 32	0.819 27	0.632 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 Net	0.690 20	0.884 7	0.754 28	0.795 20	0.647 25	0.818 26	0.422 43	0.802 20	0.612 17	0.604 18	0.945 36	0.462 31	0.189 53	0.563 37	0.853 11	0.726 10	0.765 23	0.632 15	0.904 15	0.821 25	0.606 37
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
Feature_GeometricNet	0.690 20	0.884 7	0.754 28	0.795 20	0.647 25	0.818 26	0.422 43	0.802 20	0.612 17	0.604 18	0.945 36	0.462 31	0.189 53	0.563 37	0.853 11	0.726 10	0.765 23	0.632 15	0.904 15	0.821 25	0.606 37
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
PointConv	0.666 30	0.781 25	0.759 24	0.699 42	0.644 28	0.822 22	0.475 21	0.779 23	0.564 37	0.504 45	0.953 15	0.428 47	0.203 49	0.586 30	0.754 29	0.661 33	0.753 29	0.588 35	0.902 18	0.813 30	0.642 24
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointContrast_LA_SEM	0.683 25	0.757 36	0.784 17	0.786 23	0.639 29	0.824 21	0.408 48	0.775 24	0.604 23	0.541 32	0.934 58	0.532 12	0.269 27	0.552 40	0.777 26	0.645 41	0.793 12	0.640 12	0.913 12	0.824 22	0.671 16

ROSMRF3D	0.673 28	0.789 23	0.748 31	0.763 31	0.635 30	0.814 30	0.407 50	0.747 32	0.581 30	0.573 25	0.950 23	0.484 22	0.271 26	0.607 25	0.754 29	0.649 37	0.774 19	0.596 29	0.883 31	0.823 23	0.606 37

joint point-based	0.634 43	0.614 62	0.778 19	0.667 53	0.633 31	0.825 20	0.420 45	0.804 18	0.467 60	0.561 27	0.951 19	0.494 20	0.291 18	0.566 36	0.458 60	0.579 56	0.764 25	0.559 44	0.838 52	0.814 28	0.598 42
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
VI-PointConv	0.676 27	0.770 32	0.754 28	0.783 25	0.621 32	0.814 30	0.552 4	0.758 28	0.571 34	0.557 28	0.954 13	0.529 13	0.268 29	0.530 46	0.682 42	0.675 27	0.719 37	0.603 27	0.888 29	0.833 18	0.665 17
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
DCM-Net	0.658 32	0.778 26	0.702 45	0.806 16	0.619 33	0.813 33	0.468 25	0.693 47	0.494 53	0.524 38	0.941 46	0.449 40	0.298 16	0.510 50	0.821 20	0.675 27	0.727 36	0.568 40	0.826 54	0.803 33	0.637 26
Jonas Schult, Francis Engelmann, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
Supervoxel-CNN	0.635 42	0.656 56	0.711 42	0.719 38	0.613 34	0.757 58	0.444 39	0.765 27	0.534 44	0.566 26	0.928 60	0.478 25	0.272 25	0.636 17	0.531 54	0.664 31	0.645 60	0.508 58	0.864 46	0.792 43	0.611 33

3DSM_DMMF	0.631 46	0.626 60	0.745 32	0.801 18	0.607 35	0.751 59	0.506 11	0.729 39	0.565 36	0.491 47	0.866 74	0.434 43	0.197 51	0.595 26	0.630 47	0.709 18	0.705 45	0.560 42	0.875 38	0.740 61	0.491 65

FPConv	0.639 39	0.785 24	0.760 23	0.713 41	0.603 36	0.798 43	0.392 55	0.534 67	0.603 24	0.524 38	0.948 28	0.457 34	0.250 33	0.538 44	0.723 35	0.598 52	0.696 48	0.614 22	0.872 42	0.799 34	0.567 52
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
HPEIN	0.618 52	0.729 44	0.668 58	0.647 59	0.597 37	0.766 55	0.414 46	0.680 49	0.520 47	0.525 37	0.946 33	0.432 44	0.215 44	0.493 55	0.599 49	0.638 42	0.617 65	0.570 38	0.897 22	0.806 31	0.605 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
PointMTL	0.632 45	0.731 43	0.688 55	0.675 48	0.591 38	0.784 49	0.444 39	0.565 65	0.610 19	0.492 46	0.949 25	0.456 35	0.254 32	0.587 28	0.706 38	0.599 51	0.665 57	0.612 25	0.868 45	0.791 46	0.579 47

MVPNet	0.641 36	0.831 16	0.715 41	0.671 51	0.590 39	0.781 50	0.394 54	0.679 50	0.642 10	0.553 29	0.937 52	0.462 31	0.256 31	0.649 15	0.406 65	0.626 45	0.691 49	0.666 9	0.877 36	0.792 43	0.608 36
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
FusionAwareConv	0.630 49	0.604 64	0.741 35	0.766 30	0.590 39	0.747 60	0.501 13	0.734 37	0.503 51	0.527 36	0.919 66	0.454 36	0.323 10	0.550 43	0.420 64	0.678 26	0.688 50	0.544 50	0.896 23	0.795 38	0.627 30
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DVVNet	0.562 60	0.648 57	0.700 47	0.770 28	0.586 41	0.687 67	0.333 65	0.650 54	0.514 49	0.475 52	0.906 70	0.359 61	0.223 41	0.340 68	0.442 63	0.422 72	0.668 56	0.501 59	0.708 68	0.779 49	0.534 59

SAFNet-seg	0.654 34	0.752 37	0.734 37	0.664 54	0.583 42	0.815 29	0.399 52	0.754 30	0.639 11	0.535 34	0.942 44	0.470 28	0.309 13	0.665 14	0.539 52	0.650 36	0.708 43	0.635 14	0.857 49	0.793 41	0.642 24
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Net	0.645 35	0.778 26	0.731 38	0.699 42	0.577 43	0.829 17	0.446 34	0.736 36	0.477 57	0.523 40	0.945 36	0.454 36	0.269 27	0.484 57	0.749 32	0.618 48	0.738 33	0.599 28	0.827 53	0.792 43	0.621 31

PointSPNet	0.637 40	0.734 42	0.692 52	0.714 40	0.576 44	0.797 44	0.446 34	0.743 34	0.598 26	0.437 58	0.942 44	0.403 54	0.150 67	0.626 21	0.800 24	0.649 37	0.697 47	0.557 45	0.846 51	0.777 51	0.563 53

MCCNN	0.633 44	0.866 11	0.731 38	0.771 27	0.576 44	0.809 34	0.410 47	0.684 48	0.497 52	0.491 47	0.949 25	0.466 30	0.105 72	0.581 31	0.646 46	0.620 46	0.680 52	0.542 52	0.817 56	0.795 38	0.618 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
PointMRNet	0.640 38	0.717 48	0.701 46	0.692 44	0.576 44	0.801 39	0.467 27	0.716 41	0.563 38	0.459 54	0.953 15	0.429 46	0.169 60	0.581 31	0.854 10	0.605 49	0.710 40	0.550 48	0.894 25	0.793 41	0.575 48

SConv	0.636 41	0.830 17	0.697 49	0.752 34	0.572 47	0.780 51	0.445 36	0.716 41	0.529 45	0.530 35	0.951 19	0.446 42	0.170 59	0.507 52	0.666 44	0.636 43	0.682 51	0.541 53	0.886 30	0.799 34	0.594 44

HPGCNN	0.656 33	0.698 52	0.743 34	0.650 57	0.564 48	0.820 25	0.505 12	0.758 28	0.631 13	0.479 50	0.945 36	0.480 24	0.226 39	0.572 34	0.774 27	0.690 23	0.735 35	0.614 22	0.853 50	0.776 52	0.597 43
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
ROSMRF	0.580 58	0.772 29	0.707 43	0.681 47	0.563 49	0.764 56	0.362 61	0.515 68	0.465 61	0.465 53	0.936 53	0.427 49	0.207 46	0.438 60	0.577 50	0.536 61	0.675 54	0.486 63	0.723 67	0.779 49	0.524 61

SIConv	0.625 51	0.830 17	0.694 50	0.757 32	0.563 49	0.772 54	0.448 33	0.647 56	0.520 47	0.509 42	0.949 25	0.431 45	0.191 52	0.496 54	0.614 48	0.647 40	0.672 55	0.535 55	0.876 37	0.783 48	0.571 49

PointConv-SFPN	0.641 36	0.776 28	0.703 44	0.721 37	0.557 51	0.826 19	0.451 31	0.672 52	0.563 38	0.483 49	0.943 43	0.425 50	0.162 63	0.644 16	0.726 34	0.659 34	0.709 42	0.572 37	0.875 38	0.786 47	0.559 55

APCF-Net	0.631 46	0.742 40	0.687 57	0.672 49	0.557 51	0.792 47	0.408 48	0.665 53	0.545 41	0.508 43	0.952 18	0.428 47	0.186 55	0.634 19	0.702 39	0.620 46	0.706 44	0.555 46	0.873 41	0.798 36	0.581 46
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
DenSeR	0.628 50	0.800 22	0.625 68	0.719 38	0.545 53	0.806 36	0.445 36	0.597 60	0.448 64	0.519 41	0.938 51	0.481 23	0.328 8	0.489 56	0.499 59	0.657 35	0.759 27	0.592 33	0.881 34	0.797 37	0.634 27

SPH3D-GCN	0.610 53	0.858 13	0.772 20	0.489 73	0.532 54	0.792 47	0.404 51	0.643 57	0.570 35	0.507 44	0.935 54	0.414 53	0.046 78	0.510 50	0.702 39	0.602 50	0.705 45	0.549 49	0.859 48	0.773 53	0.534 59
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
PointNet2-SFPN	0.631 46	0.771 30	0.692 52	0.672 49	0.524 55	0.837 12	0.440 41	0.706 45	0.538 43	0.446 56	0.944 41	0.421 51	0.219 42	0.552 40	0.751 31	0.591 53	0.737 34	0.543 51	0.901 20	0.768 54	0.557 56

AttAN	0.609 54	0.760 34	0.667 59	0.649 58	0.521 56	0.793 45	0.457 30	0.648 55	0.528 46	0.434 60	0.947 31	0.401 55	0.153 66	0.454 59	0.721 36	0.648 39	0.717 38	0.536 54	0.904 15	0.765 55	0.485 66
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
TextureNet	0.566 59	0.672 55	0.664 60	0.671 51	0.494 57	0.719 63	0.445 36	0.678 51	0.411 70	0.396 63	0.935 54	0.356 62	0.225 40	0.412 64	0.535 53	0.565 58	0.636 64	0.464 66	0.794 60	0.680 70	0.568 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++ & Feature	0.557 61	0.735 41	0.661 61	0.686 45	0.491 58	0.744 61	0.392 55	0.539 66	0.451 63	0.375 65	0.946 33	0.376 58	0.205 47	0.403 65	0.356 68	0.553 60	0.643 61	0.497 60	0.824 55	0.756 57	0.515 62

DPC	0.592 56	0.720 46	0.700 47	0.602 66	0.480 59	0.762 57	0.380 60	0.713 43	0.585 29	0.437 58	0.940 48	0.369 59	0.288 19	0.434 62	0.509 58	0.590 55	0.639 63	0.567 41	0.772 62	0.755 58	0.592 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, FTSDF	0.501 67	0.558 69	0.608 71	0.424 77	0.478 60	0.690 66	0.246 73	0.586 63	0.468 59	0.450 55	0.911 68	0.394 56	0.160 64	0.438 60	0.212 74	0.432 71	0.541 72	0.475 65	0.742 65	0.727 63	0.477 67

GMLPs	0.538 63	0.495 73	0.693 51	0.647 59	0.471 61	0.793 45	0.300 67	0.477 69	0.505 50	0.358 67	0.903 72	0.327 66	0.081 75	0.472 58	0.529 55	0.448 70	0.710 40	0.509 56	0.746 64	0.737 62	0.554 57

CCRFNet	0.589 57	0.766 33	0.659 62	0.683 46	0.470 62	0.740 62	0.387 59	0.620 59	0.490 54	0.476 51	0.922 64	0.355 63	0.245 36	0.511 49	0.511 57	0.571 57	0.643 61	0.493 62	0.872 42	0.762 56	0.600 41

LAP-D	0.594 55	0.720 46	0.692 52	0.637 62	0.456 63	0.773 53	0.391 57	0.730 38	0.587 27	0.445 57	0.940 48	0.381 57	0.288 19	0.434 62	0.453 61	0.591 53	0.649 58	0.581 36	0.777 61	0.749 60	0.610 35

subcloud_weak	0.516 65	0.676 54	0.591 73	0.609 63	0.442 64	0.774 52	0.335 64	0.597 60	0.422 69	0.357 68	0.932 59	0.341 65	0.094 74	0.298 70	0.528 56	0.473 68	0.676 53	0.495 61	0.602 74	0.721 64	0.349 75

Online SegFusion	0.515 66	0.607 63	0.644 66	0.579 68	0.434 65	0.630 73	0.353 62	0.628 58	0.440 65	0.410 61	0.762 78	0.307 68	0.167 61	0.520 47	0.403 66	0.516 62	0.565 67	0.447 70	0.678 70	0.701 67	0.514 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-lite	0.553 62	0.633 59	0.648 63	0.659 55	0.430 66	0.800 40	0.390 58	0.592 62	0.454 62	0.371 66	0.939 50	0.368 60	0.136 70	0.368 66	0.448 62	0.560 59	0.715 39	0.486 63	0.882 32	0.720 65	0.462 69

3DMV	0.484 69	0.484 75	0.538 75	0.643 61	0.424 67	0.606 76	0.310 66	0.574 64	0.433 68	0.378 64	0.796 76	0.301 69	0.214 45	0.537 45	0.208 75	0.472 69	0.507 76	0.413 74	0.693 69	0.602 75	0.539 58
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PCNN	0.498 68	0.559 68	0.644 66	0.560 70	0.420 68	0.711 65	0.229 75	0.414 70	0.436 66	0.352 69	0.941 46	0.324 67	0.155 65	0.238 74	0.387 67	0.493 64	0.529 73	0.509 56	0.813 58	0.751 59	0.504 64

PanopticFusion-label	0.529 64	0.491 74	0.688 55	0.604 65	0.386 69	0.632 72	0.225 77	0.705 46	0.434 67	0.293 72	0.815 75	0.348 64	0.241 37	0.499 53	0.669 43	0.507 63	0.649 58	0.442 71	0.796 59	0.602 75	0.561 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)
FCPN	0.447 71	0.679 53	0.604 72	0.578 69	0.380 70	0.682 68	0.291 70	0.106 79	0.483 56	0.258 77	0.920 65	0.258 74	0.025 79	0.231 76	0.325 69	0.480 67	0.560 69	0.463 67	0.725 66	0.666 72	0.231 79
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
Tangent Convolutions	0.438 74	0.437 77	0.646 65	0.474 74	0.369 71	0.645 71	0.353 62	0.258 76	0.282 78	0.279 73	0.918 67	0.298 70	0.147 69	0.283 71	0.294 70	0.487 65	0.562 68	0.427 73	0.619 73	0.633 73	0.352 74
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
PNET2	0.442 72	0.548 70	0.548 74	0.597 67	0.363 72	0.628 74	0.300 67	0.292 74	0.374 72	0.307 71	0.881 73	0.268 73	0.186 55	0.238 74	0.204 76	0.407 73	0.506 77	0.449 69	0.667 71	0.620 74	0.462 69

ScanNet+FTSDF	0.383 77	0.297 79	0.491 77	0.432 76	0.358 73	0.612 75	0.274 71	0.116 78	0.411 70	0.265 75	0.904 71	0.229 76	0.079 76	0.250 72	0.185 77	0.320 77	0.510 74	0.385 76	0.548 76	0.597 78	0.394 72

SurfaceConvPF	0.442 72	0.505 72	0.622 69	0.380 78	0.342 74	0.654 70	0.227 76	0.397 72	0.367 73	0.276 74	0.924 62	0.240 75	0.198 50	0.359 67	0.262 71	0.366 74	0.581 66	0.435 72	0.640 72	0.668 71	0.398 71
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
3DWSSS	0.425 75	0.525 71	0.647 64	0.522 71	0.324 75	0.488 79	0.077 80	0.712 44	0.353 74	0.401 62	0.636 80	0.281 72	0.176 58	0.340 68	0.565 51	0.175 80	0.551 70	0.398 75	0.370 80	0.602 75	0.361 73

PointCNN with RGB	0.458 70	0.577 67	0.611 70	0.356 79	0.321 76	0.715 64	0.299 69	0.376 73	0.328 76	0.319 70	0.944 41	0.285 71	0.164 62	0.216 77	0.229 73	0.484 66	0.545 71	0.456 68	0.755 63	0.709 66	0.475 68
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
SPLAT Net	0.393 76	0.472 76	0.511 76	0.606 64	0.311 77	0.656 69	0.245 74	0.405 71	0.328 76	0.197 78	0.927 61	0.227 77	0.000 81	0.001 81	0.249 72	0.271 79	0.510 74	0.383 77	0.593 75	0.699 68	0.267 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
ScanNet	0.306 80	0.203 80	0.366 79	0.501 72	0.311 77	0.524 78	0.211 78	0.002 81	0.342 75	0.189 79	0.786 77	0.145 80	0.102 73	0.245 73	0.152 78	0.318 78	0.348 79	0.300 79	0.460 79	0.437 80	0.182 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++	0.339 78	0.584 66	0.478 78	0.458 75	0.256 79	0.360 80	0.250 72	0.247 77	0.278 79	0.261 76	0.677 79	0.183 78	0.117 71	0.212 78	0.145 79	0.364 75	0.346 80	0.232 80	0.548 76	0.523 79	0.252 78
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNet	0.308 79	0.353 78	0.290 80	0.278 80	0.166 80	0.553 77	0.169 79	0.286 75	0.147 80	0.148 80	0.908 69	0.182 79	0.064 77	0.023 80	0.018 81	0.354 76	0.363 78	0.345 78	0.546 78	0.685 69	0.278 76

ERROR	0.054 81	0.000 81	0.041 81	0.172 81	0.030 81	0.062 81	0.001 81	0.035 80	0.004 81	0.051 81	0.143 81	0.019 81	0.003 80	0.041 79	0.050 80	0.003 81	0.054 81	0.018 81	0.005 81	0.264 81	0.082 81

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

Method	avg ap 50%	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	window

SoftGroup	0.761 1	1.000 1	0.808 9	0.845 4	0.716 1	0.862 3	0.243 6	0.824 2	0.655 3	0.620 1	0.734 1	0.699 2	0.791 2	0.981 15	0.716 3	0.844 1	0.769 1	1.000 1	0.594 3

IPCA-Inst	0.731 2	1.000 1	0.788 13	0.884 3	0.698 2	0.788 15	0.252 5	0.760 8	0.646 4	0.511 6	0.637 2	0.665 3	0.804 1	1.000 1	0.644 9	0.778 5	0.747 2	1.000 1	0.561 6

HAIS	0.699 3	1.000 1	0.849 3	0.820 5	0.675 3	0.808 9	0.279 4	0.757 9	0.465 9	0.517 5	0.596 3	0.559 5	0.600 13	1.000 1	0.654 8	0.767 6	0.676 6	0.994 17	0.560 7
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
SphereSeg	0.680 5	1.000 1	0.856 2	0.744 11	0.618 4	0.893 1	0.151 9	0.651 14	0.713 1	0.537 4	0.579 7	0.430 14	0.651 4	1.000 1	0.389 24	0.744 13	0.697 3	0.991 18	0.601 2

Mask-Group	0.664 7	1.000 1	0.822 6	0.764 10	0.616 5	0.815 6	0.139 13	0.694 12	0.597 5	0.459 11	0.566 8	0.599 4	0.600 13	0.516 29	0.715 4	0.819 3	0.635 10	1.000 1	0.603 1

DENet	0.629 16	1.000 1	0.797 11	0.608 17	0.589 6	0.627 23	0.219 7	0.882 1	0.310 20	0.402 22	0.383 25	0.396 18	0.650 5	1.000 1	0.663 6	0.543 29	0.691 5	1.000 1	0.568 5

INS-Conv-instance	0.657 8	1.000 1	0.760 17	0.667 14	0.581 7	0.863 2	0.323 3	0.655 13	0.477 7	0.473 9	0.549 10	0.432 13	0.650 5	1.000 1	0.655 7	0.738 14	0.585 15	0.944 22	0.472 15

RWSeg	0.567 20	0.528 31	0.708 24	0.626 15	0.580 8	0.745 19	0.063 22	0.627 15	0.240 23	0.400 23	0.497 15	0.464 9	0.515 20	1.000 1	0.475 18	0.745 12	0.571 16	1.000 1	0.429 19

OccuSeg+instance	0.672 6	1.000 1	0.758 19	0.682 13	0.576 9	0.842 4	0.477 1	0.504 25	0.524 6	0.567 2	0.585 6	0.451 10	0.557 18	1.000 1	0.751 2	0.797 4	0.563 18	1.000 1	0.467 16

DD-UNet+Group	0.635 15	0.667 25	0.797 12	0.714 12	0.562 10	0.774 17	0.146 10	0.810 5	0.429 12	0.476 8	0.546 12	0.399 17	0.633 12	1.000 1	0.632 10	0.722 16	0.609 12	1.000 1	0.514 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
SSTNet	0.698 4	1.000 1	0.697 25	0.888 2	0.556 11	0.803 10	0.387 2	0.626 16	0.417 13	0.556 3	0.585 5	0.702 1	0.600 13	1.000 1	0.824 1	0.720 17	0.692 4	1.000 1	0.509 10
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
RPGN	0.643 11	1.000 1	0.758 18	0.582 22	0.539 12	0.826 5	0.046 24	0.765 7	0.372 16	0.436 16	0.588 4	0.539 7	0.650 5	1.000 1	0.577 12	0.750 11	0.653 9	0.997 14	0.495 13

PE	0.645 10	1.000 1	0.773 14	0.798 7	0.538 13	0.786 16	0.088 20	0.799 6	0.350 18	0.435 17	0.547 11	0.545 6	0.646 11	0.933 16	0.562 14	0.761 9	0.556 22	0.997 14	0.501 12
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
Dyco3D	0.641 12	1.000 1	0.841 4	0.893 1	0.531 14	0.802 11	0.115 17	0.588 21	0.448 10	0.438 14	0.537 13	0.430 15	0.550 19	0.857 18	0.534 15	0.764 8	0.657 7	0.987 19	0.568 4
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
3D-MPA	0.611 17	1.000 1	0.833 5	0.765 9	0.526 15	0.756 18	0.136 15	0.588 21	0.470 8	0.438 15	0.432 21	0.358 21	0.650 5	0.857 18	0.429 20	0.765 7	0.557 20	1.000 1	0.430 18
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
CSC-Pretrained	0.648 9	1.000 1	0.810 7	0.768 8	0.523 16	0.813 7	0.143 12	0.819 3	0.389 14	0.422 18	0.511 14	0.443 11	0.650 5	1.000 1	0.624 11	0.732 15	0.634 11	1.000 1	0.375 22

Occipital-SCS	0.512 23	1.000 1	0.716 21	0.509 23	0.506 17	0.611 24	0.092 19	0.602 20	0.177 26	0.346 25	0.383 24	0.165 26	0.442 24	0.850 22	0.386 25	0.618 25	0.543 23	0.889 26	0.389 21

PointGroup	0.636 14	1.000 1	0.765 15	0.624 16	0.505 18	0.797 12	0.116 16	0.696 11	0.384 15	0.441 13	0.559 9	0.476 8	0.596 16	1.000 1	0.666 5	0.756 10	0.556 21	0.997 14	0.513 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]
GICN	0.638 13	1.000 1	0.895 1	0.800 6	0.480 19	0.676 20	0.144 11	0.737 10	0.354 17	0.447 12	0.400 23	0.365 20	0.700 3	1.000 1	0.569 13	0.836 2	0.599 13	1.000 1	0.473 14

SSEN	0.575 19	1.000 1	0.761 16	0.473 24	0.477 20	0.795 13	0.066 21	0.529 23	0.658 2	0.460 10	0.461 18	0.380 19	0.331 28	0.859 17	0.401 23	0.692 21	0.653 8	1.000 1	0.348 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-CNN	0.515 22	1.000 1	0.538 31	0.282 27	0.468 21	0.790 14	0.173 8	0.345 27	0.429 11	0.413 21	0.484 16	0.176 25	0.595 17	0.591 27	0.522 16	0.668 23	0.476 25	0.986 20	0.327 25

PCJC	0.578 18	1.000 1	0.810 8	0.583 21	0.449 22	0.813 8	0.042 25	0.603 19	0.341 19	0.490 7	0.465 17	0.410 16	0.650 5	0.835 23	0.264 28	0.694 20	0.561 19	0.889 26	0.504 11

MASC	0.447 27	0.528 31	0.555 29	0.381 25	0.382 23	0.633 22	0.002 31	0.509 24	0.260 22	0.361 24	0.432 20	0.327 22	0.451 22	0.571 28	0.367 26	0.639 24	0.386 26	0.980 21	0.276 27
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
R-PointNet	0.306 31	0.500 33	0.405 34	0.311 26	0.348 24	0.589 25	0.054 23	0.068 34	0.126 28	0.283 28	0.290 27	0.028 34	0.219 31	0.214 32	0.331 27	0.396 34	0.275 31	0.821 31	0.245 28

MTML	0.549 21	1.000 1	0.807 10	0.588 20	0.327 25	0.647 21	0.004 30	0.815 4	0.180 25	0.418 19	0.364 26	0.182 24	0.445 23	1.000 1	0.442 19	0.688 22	0.571 17	1.000 1	0.396 20
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
3D-BoNet	0.488 24	1.000 1	0.672 26	0.590 19	0.301 26	0.484 32	0.098 18	0.620 17	0.306 21	0.341 26	0.259 28	0.125 28	0.434 25	0.796 24	0.402 22	0.499 31	0.513 24	0.909 25	0.439 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-ResNet	0.459 26	1.000 1	0.737 20	0.159 34	0.259 27	0.587 26	0.138 14	0.475 26	0.217 24	0.416 20	0.408 22	0.128 27	0.315 29	0.714 25	0.411 21	0.536 30	0.590 14	0.873 29	0.304 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-inst	0.478 25	0.667 25	0.712 23	0.595 18	0.259 28	0.550 29	0.000 33	0.613 18	0.175 27	0.250 30	0.434 19	0.437 12	0.411 27	0.857 18	0.485 17	0.591 28	0.267 33	0.944 22	0.359 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-SIS	0.382 28	1.000 1	0.432 33	0.245 29	0.190 29	0.577 27	0.013 28	0.263 29	0.033 33	0.320 27	0.240 29	0.075 30	0.422 26	0.857 18	0.117 31	0.699 18	0.271 32	0.883 28	0.235 29
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
UNet-backbone	0.319 30	0.667 25	0.715 22	0.233 30	0.189 30	0.479 33	0.008 29	0.218 30	0.067 32	0.201 31	0.173 31	0.107 29	0.123 33	0.438 30	0.150 29	0.615 26	0.355 27	0.916 24	0.093 35

RandSA	0.250 32	0.333 34	0.613 27	0.229 31	0.163 31	0.493 30	0.000 33	0.304 28	0.107 29	0.147 33	0.100 32	0.052 33	0.231 30	0.119 33	0.039 33	0.445 33	0.325 28	0.654 32	0.141 32

Hier3D	0.323 29	0.667 25	0.542 30	0.264 28	0.157 32	0.550 28	0.000 33	0.205 32	0.009 34	0.270 29	0.218 30	0.075 30	0.500 21	0.688 26	0.007 36	0.698 19	0.301 30	0.459 34	0.200 30
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
Region	0.248 33	0.667 25	0.437 32	0.188 32	0.153 33	0.491 31	0.000 33	0.208 31	0.094 31	0.153 32	0.099 33	0.057 32	0.217 32	0.119 33	0.039 33	0.466 32	0.302 29	0.640 33	0.140 33

Sgpn_scannet	0.143 35	0.208 36	0.390 35	0.169 33	0.065 34	0.275 35	0.029 26	0.069 33	0.000 35	0.087 35	0.043 34	0.014 36	0.027 36	0.000 35	0.112 32	0.351 35	0.168 35	0.438 35	0.138 34

MaskRCNN 2d->3d Proj	0.058 36	0.333 34	0.002 36	0.000 36	0.053 35	0.002 36	0.002 32	0.021 36	0.000 35	0.045 36	0.024 36	0.238 23	0.065 35	0.000 35	0.014 35	0.107 36	0.020 36	0.110 36	0.006 36

3D-BEVIS	0.248 33	0.667 25	0.566 28	0.076 35	0.035 36	0.394 34	0.027 27	0.035 35	0.098 30	0.099 34	0.030 35	0.025 35	0.098 34	0.375 31	0.126 30	0.604 27	0.181 34	0.854 30	0.171 31
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.

Method	avg iou	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	floor	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	wall	window

Virtual MVFusion (R)	0.745 1	0.861 1	0.839 1	0.881 1	0.672 1	0.512 1	0.422 10	0.898 1	0.723 1	0.714 1	0.954 2	0.454 1	0.509 1	0.773 1	0.895 1	0.756 1	0.820 1	0.653 1	0.935 1	0.891 1	0.728 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_2D	0.670 2	0.822 3	0.795 3	0.836 2	0.659 2	0.481 2	0.451 6	0.769 2	0.656 3	0.567 3	0.931 3	0.395 3	0.390 4	0.700 2	0.534 3	0.689 5	0.770 2	0.574 3	0.865 3	0.831 3	0.675 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 Net	0.636 3	0.825 2	0.820 2	0.179 16	0.648 3	0.463 3	0.549 1	0.742 3	0.676 2	0.628 2	0.961 1	0.420 2	0.379 5	0.684 3	0.381 10	0.732 2	0.723 3	0.599 2	0.827 8	0.851 2	0.634 3

DMMF_3d	0.605 4	0.651 6	0.744 7	0.782 3	0.637 4	0.387 4	0.536 2	0.732 4	0.590 5	0.540 4	0.856 13	0.359 6	0.306 10	0.596 6	0.539 2	0.627 12	0.706 4	0.497 6	0.785 12	0.757 11	0.476 13

MCA-Net	0.595 6	0.533 12	0.756 5	0.746 5	0.590 5	0.334 8	0.506 3	0.670 7	0.587 6	0.500 8	0.905 8	0.366 5	0.352 6	0.601 5	0.506 6	0.669 10	0.648 6	0.501 5	0.839 7	0.769 9	0.516 12

DMMF	0.597 5	0.543 11	0.755 6	0.749 4	0.585 6	0.338 6	0.494 4	0.704 6	0.598 4	0.494 10	0.911 6	0.347 8	0.327 9	0.593 7	0.527 4	0.675 7	0.646 8	0.513 4	0.842 6	0.774 8	0.527 11

RFBNet	0.592 7	0.616 7	0.758 4	0.659 6	0.581 7	0.330 9	0.469 5	0.655 10	0.543 9	0.524 5	0.924 4	0.355 7	0.336 8	0.572 8	0.479 8	0.671 8	0.648 6	0.480 7	0.814 10	0.814 4	0.614 6

SSMA	0.577 9	0.695 4	0.716 10	0.439 12	0.563 8	0.314 10	0.444 8	0.719 5	0.551 7	0.503 7	0.887 10	0.346 9	0.348 7	0.603 4	0.353 12	0.709 3	0.600 11	0.457 10	0.901 2	0.786 6	0.599 7
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019
SN_RN152pyrx8_RVC	0.546 10	0.572 9	0.663 13	0.638 8	0.518 9	0.298 12	0.366 15	0.633 12	0.510 11	0.446 12	0.864 11	0.296 11	0.267 12	0.542 10	0.346 13	0.704 4	0.575 14	0.431 12	0.853 5	0.766 10	0.630 4

FuseNet	0.535 11	0.570 10	0.681 12	0.182 15	0.512 10	0.290 13	0.431 9	0.659 9	0.504 12	0.495 9	0.903 9	0.308 10	0.428 3	0.523 12	0.365 11	0.676 6	0.621 10	0.470 9	0.762 13	0.779 7	0.541 9
Caner Hazirbas, Lingni Ma, Csaba Domokos, Daniel Cremers: FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-based CNN Architecture. ACCV 2016
DCRedNet	0.583 8	0.682 5	0.723 8	0.542 10	0.510 11	0.310 11	0.451 6	0.668 8	0.549 8	0.520 6	0.920 5	0.375 4	0.446 2	0.528 11	0.417 9	0.670 9	0.577 13	0.478 8	0.862 4	0.806 5	0.628 5

ILC-PSPNet	0.475 15	0.490 14	0.581 15	0.289 14	0.507 12	0.067 17	0.379 13	0.610 14	0.417 15	0.435 13	0.822 16	0.278 12	0.267 12	0.503 13	0.228 14	0.616 14	0.533 15	0.375 14	0.820 9	0.729 12	0.560 8

3DMV (2d proj)	0.498 13	0.481 15	0.612 14	0.579 9	0.456 13	0.343 5	0.384 12	0.623 13	0.525 10	0.381 14	0.845 14	0.254 13	0.264 14	0.557 9	0.182 15	0.581 15	0.598 12	0.429 13	0.760 14	0.661 16	0.446 15
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
MSeg1080_RVC	0.485 14	0.505 13	0.709 11	0.092 17	0.427 14	0.241 14	0.411 11	0.654 11	0.385 17	0.457 11	0.861 12	0.053 17	0.279 11	0.503 13	0.481 7	0.645 11	0.626 9	0.365 15	0.748 15	0.725 13	0.529 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 16	0.264 17	0.452 17	0.452 11	0.365 15	0.181 15	0.143 17	0.456 16	0.409 16	0.346 16	0.769 17	0.164 15	0.218 15	0.359 16	0.123 17	0.403 17	0.381 17	0.313 17	0.571 16	0.685 15	0.472 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)	0.330 17	0.293 16	0.521 16	0.657 7	0.361 16	0.161 16	0.250 16	0.004 17	0.440 14	0.183 17	0.836 15	0.125 16	0.060 17	0.319 17	0.132 16	0.417 16	0.412 16	0.344 16	0.541 17	0.427 17	0.109 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++	0.503 12	0.613 8	0.722 9	0.418 13	0.358 17	0.337 7	0.370 14	0.479 15	0.443 13	0.368 15	0.907 7	0.207 14	0.213 16	0.464 15	0.525 5	0.618 13	0.657 5	0.450 11	0.788 11	0.721 14	0.408 16
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019

Method	avg ap	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	window

UniDet_RVC	0.205 1	0.381 1	0.323 1	0.037 1	0.226 1	0.177 1	0.063 1	0.277 1	0.120 1	0.067 1	0.131 1	0.074 2	0.317 1	0.080 1	0.235 1	0.289 1	0.141 1	0.678 1	0.080 1

MaskRCNN_ScanNet	0.119 2	0.129 2	0.212 2	0.002 2	0.112 2	0.148 2	0.014 2	0.205 2	0.044 2	0.066 2	0.078 2	0.095 1	0.142 2	0.030 2	0.128 2	0.139 2	0.080 2	0.459 2	0.057 2
Re-implementation of Kaiming He, Georgia Gkioxari, Piotr Dollár, Ross Girshick: Mask R-CNN. ICCV'17

Method	avg recall	apartment	bathroom	bedroom / hotel	bookstore / library	conference room	copy/mail room	hallway	kitchen	laundry room	living room / lounge	misc	office	storage / basement / garage

SE-ResNeXt-SSMA	0.498 3	0.000 4	0.812 3	0.941 1	0.500 2	0.500 3	0.500 2	0.500 2	0.429 4	0.500 2	0.667 2	0.500 1	0.625 3	0.000 2
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. arXiv
multi-task	0.700 1	0.500 1	1.000 1	0.882 2	0.500 2	1.000 1	1.000 1	0.500 2	1.000 1	1.000 1	0.778 1	0.000 2	0.938 1	0.000 2
Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler: Indoor Scene Recognition in 3D. IROS 2020
3DASPP-SCE	0.691 2	0.500 1	0.938 2	0.824 3	1.000 1	1.000 1	0.500 2	1.000 1	0.857 2	0.500 2	0.556 3	0.000 2	0.812 2	0.500 1

resnet50_scannet	0.353 4	0.250 3	0.812 3	0.529 4	0.500 2	0.500 3	0.000 4	0.500 2	0.571 3	0.000 4	0.556 3	0.000 2	0.375 4	0.000 2

3D Semantic label benchmark

3D Semantic instance benchmark

2D Semantic label benchmark

2D Semantic instance benchmark

Scene type classification benchmark