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

BPNet	0.749 2	0.909 1	0.818 4	0.811 8	0.752 1	0.839 6	0.485 12	0.842 7	0.673 3	0.644 4	0.957 2	0.528 6	0.305 8	0.773 2	0.859 4	0.788 2	0.818 3	0.693 2	0.916 6	0.856 5	0.723 4

OccuSeg+Semantic	0.764 1	0.758 26	0.796 7	0.839 4	0.746 2	0.907 1	0.562 1	0.850 5	0.680 2	0.672 1	0.978 1	0.610 1	0.335 2	0.777 1	0.819 12	0.847 1	0.830 1	0.691 3	0.972 1	0.885 1	0.727 2

MatchingNet	0.724 7	0.812 15	0.812 6	0.810 9	0.735 3	0.834 8	0.495 10	0.860 3	0.572 21	0.602 9	0.954 5	0.512 8	0.280 15	0.757 4	0.845 10	0.725 5	0.780 8	0.606 17	0.937 3	0.851 7	0.700 7

VMNet	0.746 3	0.870 4	0.838 1	0.858 1	0.729 4	0.850 4	0.501 7	0.874 2	0.587 17	0.658 3	0.956 3	0.564 4	0.299 9	0.765 3	0.900 1	0.716 7	0.812 4	0.631 9	0.939 2	0.858 4	0.709 5

SparseConvNet	0.725 6	0.647 43	0.821 2	0.846 3	0.721 5	0.869 2	0.533 2	0.754 16	0.603 14	0.614 6	0.955 4	0.572 3	0.325 4	0.710 7	0.870 3	0.724 6	0.823 2	0.628 10	0.934 5	0.865 3	0.683 8

MinkowskiNet	0.736 5	0.859 6	0.818 4	0.832 5	0.709 6	0.840 5	0.521 4	0.853 4	0.660 4	0.643 5	0.951 9	0.544 5	0.286 14	0.731 6	0.893 2	0.675 15	0.772 10	0.683 4	0.874 22	0.852 6	0.727 2
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
Virtual MVFusion	0.746 3	0.771 22	0.819 3	0.848 2	0.702 7	0.865 3	0.397 39	0.899 1	0.699 1	0.664 2	0.948 18	0.588 2	0.330 3	0.746 5	0.851 7	0.764 3	0.796 6	0.704 1	0.935 4	0.866 2	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
CU-Hybrid Net	0.693 10	0.596 48	0.789 8	0.803 11	0.677 8	0.800 24	0.469 16	0.846 6	0.554 29	0.591 11	0.948 18	0.500 9	0.316 6	0.609 17	0.847 9	0.732 4	0.808 5	0.593 19	0.894 12	0.839 8	0.652 11

RFCR	0.702 8	0.889 2	0.745 17	0.813 7	0.672 9	0.818 16	0.493 11	0.815 8	0.623 9	0.610 7	0.947 21	0.470 16	0.249 23	0.594 19	0.848 8	0.705 10	0.779 9	0.646 7	0.892 14	0.823 10	0.611 20

JSENet	0.699 9	0.881 3	0.762 13	0.821 6	0.667 10	0.800 24	0.522 3	0.792 10	0.613 10	0.607 8	0.935 40	0.492 11	0.205 33	0.576 25	0.853 6	0.691 11	0.758 13	0.652 6	0.872 25	0.828 9	0.649 12
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 11	0.704 36	0.741 20	0.754 19	0.656 11	0.829 10	0.501 7	0.741 20	0.609 12	0.548 15	0.950 13	0.522 7	0.371 1	0.633 14	0.756 17	0.715 8	0.771 11	0.623 11	0.861 31	0.814 12	0.658 10
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
PointASNL	0.666 14	0.703 37	0.781 9	0.751 21	0.655 12	0.830 9	0.471 15	0.769 13	0.474 44	0.537 17	0.951 9	0.475 14	0.279 16	0.635 12	0.698 29	0.675 15	0.751 15	0.553 32	0.816 41	0.806 15	0.703 6
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
SALANet	0.670 13	0.816 14	0.770 12	0.768 16	0.652 13	0.807 21	0.451 20	0.747 18	0.659 5	0.545 16	0.924 46	0.473 15	0.149 51	0.571 27	0.811 13	0.635 27	0.746 16	0.623 11	0.892 14	0.794 24	0.570 34

KP-FCNN	0.684 12	0.847 8	0.758 16	0.784 13	0.647 14	0.814 18	0.473 14	0.772 12	0.605 13	0.594 10	0.935 40	0.450 25	0.181 41	0.587 20	0.805 14	0.690 12	0.785 7	0.614 13	0.882 17	0.819 11	0.632 16
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
PointConv	0.666 14	0.781 17	0.759 15	0.699 26	0.644 15	0.822 14	0.475 13	0.779 11	0.564 25	0.504 29	0.953 6	0.428 33	0.203 35	0.586 22	0.754 18	0.661 19	0.753 14	0.588 20	0.902 8	0.813 14	0.642 13
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
joint point-based	0.634 28	0.614 46	0.778 10	0.667 39	0.633 16	0.825 13	0.420 32	0.804 9	0.467 46	0.561 13	0.951 9	0.494 10	0.291 11	0.566 28	0.458 43	0.579 42	0.764 12	0.559 29	0.838 36	0.814 12	0.598 26
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
DCM-Net	0.658 16	0.778 18	0.702 31	0.806 10	0.619 17	0.813 19	0.468 17	0.693 31	0.494 40	0.524 23	0.941 33	0.449 26	0.298 10	0.510 35	0.821 11	0.675 15	0.727 21	0.568 25	0.826 38	0.803 17	0.637 15
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 27	0.656 41	0.711 27	0.719 23	0.613 18	0.757 41	0.444 28	0.765 14	0.534 32	0.566 12	0.928 44	0.478 13	0.272 17	0.636 11	0.531 39	0.664 18	0.645 43	0.508 43	0.864 30	0.792 27	0.611 20

3DSM_DMMF	0.631 31	0.626 45	0.745 17	0.801 12	0.607 19	0.751 42	0.506 5	0.729 24	0.565 24	0.491 32	0.866 57	0.434 29	0.197 37	0.595 18	0.630 33	0.709 9	0.705 27	0.560 28	0.875 20	0.740 47	0.491 49

FPConv	0.639 23	0.785 16	0.760 14	0.713 25	0.603 20	0.798 26	0.392 41	0.534 49	0.603 14	0.524 23	0.948 18	0.457 21	0.250 22	0.538 31	0.723 22	0.598 37	0.696 31	0.614 13	0.872 25	0.799 19	0.567 36
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 37	0.729 32	0.668 44	0.647 43	0.597 21	0.766 38	0.414 33	0.680 33	0.520 35	0.525 22	0.946 23	0.432 30	0.215 30	0.493 41	0.599 35	0.638 25	0.617 48	0.570 23	0.897 10	0.806 15	0.605 24
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
VACNN++	0.638 24	0.820 13	0.701 32	0.687 30	0.594 22	0.791 31	0.430 31	0.587 44	0.569 23	0.529 20	0.950 13	0.467 18	0.253 21	0.524 33	0.722 23	0.618 31	0.694 32	0.570 23	0.793 45	0.802 18	0.659 9

PointMTL	0.632 30	0.731 31	0.688 41	0.675 34	0.591 23	0.784 33	0.444 28	0.565 47	0.610 11	0.492 31	0.949 15	0.456 22	0.254 20	0.587 20	0.706 26	0.599 36	0.665 40	0.612 16	0.868 29	0.791 30	0.579 31

MVPNet	0.641 20	0.831 10	0.715 26	0.671 37	0.590 24	0.781 34	0.394 40	0.679 34	0.642 6	0.553 14	0.937 37	0.462 20	0.256 19	0.649 9	0.406 48	0.626 28	0.691 33	0.666 5	0.877 18	0.792 27	0.608 23
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
FusionAwareConv	0.630 34	0.604 47	0.741 20	0.766 17	0.590 24	0.747 43	0.501 7	0.734 22	0.503 38	0.527 21	0.919 50	0.454 23	0.323 5	0.550 30	0.420 47	0.678 14	0.688 34	0.544 35	0.896 11	0.795 22	0.627 17
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DVVNet	0.562 46	0.648 42	0.700 34	0.770 15	0.586 26	0.687 50	0.333 48	0.650 39	0.514 37	0.475 38	0.906 54	0.359 46	0.223 28	0.340 51	0.442 46	0.422 54	0.668 39	0.501 44	0.708 52	0.779 34	0.534 43

SAFNet-seg	0.654 18	0.752 27	0.734 22	0.664 40	0.583 27	0.815 17	0.399 38	0.754 16	0.639 7	0.535 18	0.942 31	0.470 16	0.309 7	0.665 8	0.539 37	0.650 21	0.708 25	0.635 8	0.857 33	0.793 25	0.642 13

PointMRNet-lite	0.625 35	0.643 44	0.711 27	0.697 28	0.581 28	0.801 22	0.408 35	0.670 37	0.558 28	0.497 30	0.944 27	0.436 28	0.152 49	0.617 16	0.708 25	0.603 34	0.743 17	0.532 41	0.870 28	0.784 32	0.545 41

RandLA-Net	0.645 19	0.778 18	0.731 23	0.699 26	0.577 29	0.829 10	0.446 24	0.736 21	0.477 43	0.523 25	0.945 25	0.454 23	0.269 18	0.484 42	0.749 20	0.618 31	0.738 18	0.599 18	0.827 37	0.792 27	0.621 18

MCCNN	0.633 29	0.866 5	0.731 23	0.771 14	0.576 30	0.809 20	0.410 34	0.684 32	0.497 39	0.491 32	0.949 15	0.466 19	0.105 55	0.581 23	0.646 32	0.620 29	0.680 36	0.542 37	0.817 40	0.795 22	0.618 19
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
PointSPNet	0.637 25	0.734 30	0.692 38	0.714 24	0.576 30	0.797 27	0.446 24	0.743 19	0.598 16	0.437 44	0.942 31	0.403 39	0.150 50	0.626 15	0.800 15	0.649 22	0.697 30	0.557 30	0.846 35	0.777 36	0.563 37

PointMRNet	0.640 22	0.717 35	0.701 32	0.692 29	0.576 30	0.801 22	0.467 18	0.716 26	0.563 26	0.459 40	0.953 6	0.429 32	0.169 43	0.581 23	0.854 5	0.605 33	0.710 23	0.550 33	0.894 12	0.793 25	0.575 32

SConv	0.636 26	0.830 11	0.697 36	0.752 20	0.572 33	0.780 35	0.445 26	0.716 26	0.529 33	0.530 19	0.951 9	0.446 27	0.170 42	0.507 37	0.666 31	0.636 26	0.682 35	0.541 38	0.886 16	0.799 19	0.594 28

HPGCNN	0.656 17	0.698 38	0.743 19	0.650 41	0.564 34	0.820 15	0.505 6	0.758 15	0.631 8	0.479 36	0.945 25	0.480 12	0.226 26	0.572 26	0.774 16	0.690 12	0.735 20	0.614 13	0.853 34	0.776 37	0.597 27
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
ROSMRF	0.580 44	0.772 21	0.707 29	0.681 33	0.563 35	0.764 39	0.362 46	0.515 50	0.465 47	0.465 39	0.936 38	0.427 35	0.207 32	0.438 44	0.577 36	0.536 46	0.675 37	0.486 47	0.723 51	0.779 34	0.524 45

SIConv	0.625 35	0.830 11	0.694 37	0.757 18	0.563 35	0.772 37	0.448 22	0.647 41	0.520 35	0.509 26	0.949 15	0.431 31	0.191 38	0.496 40	0.614 34	0.647 24	0.672 38	0.535 40	0.876 19	0.783 33	0.571 33

PointConv-SFPN	0.641 20	0.776 20	0.703 30	0.721 22	0.557 37	0.826 12	0.451 20	0.672 36	0.563 26	0.483 35	0.943 30	0.425 36	0.162 45	0.644 10	0.726 21	0.659 20	0.709 24	0.572 22	0.875 20	0.786 31	0.559 39

APCF-Net	0.631 31	0.742 28	0.687 43	0.672 35	0.557 37	0.792 29	0.408 35	0.665 38	0.545 30	0.508 27	0.952 8	0.428 33	0.186 39	0.634 13	0.702 27	0.620 29	0.706 26	0.555 31	0.873 24	0.798 21	0.581 30
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
SPH3D-GCN	0.610 38	0.858 7	0.772 11	0.489 54	0.532 39	0.792 29	0.404 37	0.643 42	0.570 22	0.507 28	0.935 40	0.414 38	0.046 60	0.510 35	0.702 27	0.602 35	0.705 27	0.549 34	0.859 32	0.773 38	0.534 43
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
PointNet2-SFPN	0.631 31	0.771 22	0.692 38	0.672 35	0.524 40	0.837 7	0.440 30	0.706 29	0.538 31	0.446 42	0.944 27	0.421 37	0.219 29	0.552 29	0.751 19	0.591 39	0.737 19	0.543 36	0.901 9	0.768 40	0.557 40

AttAN	0.609 39	0.760 25	0.667 45	0.649 42	0.521 41	0.793 28	0.457 19	0.648 40	0.528 34	0.434 46	0.947 21	0.401 40	0.153 48	0.454 43	0.721 24	0.648 23	0.717 22	0.536 39	0.904 7	0.765 41	0.485 50
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
SegGCN	0.589 42	0.833 9	0.731 23	0.539 52	0.514 42	0.789 32	0.448 22	0.467 51	0.573 20	0.484 34	0.936 38	0.396 41	0.061 59	0.501 38	0.507 42	0.594 38	0.700 29	0.563 27	0.874 22	0.771 39	0.493 48
Huan Lei, Naveed Akhtar, and Ajmal Mian: SegGCN: Efficient 3D Point Cloud Segmentation with Fuzzy Spherical Kernel. CVPR 2020
TextureNet	0.566 45	0.672 40	0.664 46	0.671 37	0.494 43	0.719 46	0.445 26	0.678 35	0.411 52	0.396 47	0.935 40	0.356 47	0.225 27	0.412 48	0.535 38	0.565 44	0.636 47	0.464 49	0.794 44	0.680 52	0.568 35
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 47	0.735 29	0.661 47	0.686 31	0.491 44	0.744 44	0.392 41	0.539 48	0.451 48	0.375 50	0.946 23	0.376 44	0.205 33	0.403 49	0.356 50	0.553 45	0.643 44	0.497 45	0.824 39	0.756 43	0.515 46

DPC	0.592 41	0.720 33	0.700 34	0.602 48	0.480 45	0.762 40	0.380 45	0.713 28	0.585 19	0.437 44	0.940 35	0.369 45	0.288 12	0.434 46	0.509 41	0.590 41	0.639 46	0.567 26	0.772 47	0.755 44	0.592 29
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 49	0.558 52	0.608 54	0.424 59	0.478 46	0.690 49	0.246 55	0.586 45	0.468 45	0.450 41	0.911 52	0.394 42	0.160 46	0.438 44	0.212 56	0.432 53	0.541 54	0.475 48	0.742 49	0.727 48	0.477 51

CCRFNet	0.589 42	0.766 24	0.659 48	0.683 32	0.470 47	0.740 45	0.387 44	0.620 43	0.490 41	0.476 37	0.922 48	0.355 48	0.245 24	0.511 34	0.511 40	0.571 43	0.643 44	0.493 46	0.872 25	0.762 42	0.600 25

LAP-D	0.594 40	0.720 33	0.692 38	0.637 45	0.456 48	0.773 36	0.391 43	0.730 23	0.587 17	0.445 43	0.940 35	0.381 43	0.288 12	0.434 46	0.453 44	0.591 39	0.649 41	0.581 21	0.777 46	0.749 46	0.610 22

3DMV	0.484 51	0.484 56	0.538 57	0.643 44	0.424 49	0.606 58	0.310 49	0.574 46	0.433 51	0.378 49	0.796 59	0.301 51	0.214 31	0.537 32	0.208 57	0.472 52	0.507 58	0.413 56	0.693 53	0.602 58	0.539 42
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PCNN	0.498 50	0.559 51	0.644 51	0.560 51	0.420 50	0.711 48	0.229 57	0.414 52	0.436 49	0.352 51	0.941 33	0.324 50	0.155 47	0.238 56	0.387 49	0.493 48	0.529 55	0.509 42	0.813 42	0.751 45	0.504 47

PanopticFusion-label	0.529 48	0.491 55	0.688 41	0.604 47	0.386 51	0.632 55	0.225 59	0.705 30	0.434 50	0.293 54	0.815 58	0.348 49	0.241 25	0.499 39	0.669 30	0.507 47	0.649 41	0.442 53	0.796 43	0.602 58	0.561 38
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 53	0.679 39	0.604 55	0.578 50	0.380 52	0.682 51	0.291 52	0.106 61	0.483 42	0.258 59	0.920 49	0.258 55	0.025 61	0.231 58	0.325 51	0.480 51	0.560 52	0.463 50	0.725 50	0.666 54	0.231 61
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 56	0.437 59	0.646 50	0.474 56	0.369 53	0.645 54	0.353 47	0.258 58	0.282 60	0.279 55	0.918 51	0.298 52	0.147 52	0.283 53	0.294 52	0.487 49	0.562 51	0.427 55	0.619 56	0.633 55	0.352 56
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
PNET2	0.442 54	0.548 53	0.548 56	0.597 49	0.363 54	0.628 56	0.300 50	0.292 56	0.374 55	0.307 53	0.881 56	0.268 54	0.186 39	0.238 56	0.204 58	0.407 55	0.506 59	0.449 52	0.667 54	0.620 56	0.462 53

ScanNet+FTSDF	0.383 59	0.297 61	0.491 59	0.432 58	0.358 55	0.612 57	0.274 53	0.116 60	0.411 52	0.265 57	0.904 55	0.229 58	0.079 57	0.250 54	0.185 59	0.320 59	0.510 56	0.385 57	0.548 58	0.597 60	0.394 55

SurfaceConvPF	0.442 54	0.505 54	0.622 52	0.380 60	0.342 56	0.654 53	0.227 58	0.397 54	0.367 56	0.276 56	0.924 46	0.240 57	0.198 36	0.359 50	0.262 53	0.366 56	0.581 50	0.435 54	0.640 55	0.668 53	0.398 54
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
PointCNN with RGB	0.458 52	0.577 50	0.611 53	0.356 61	0.321 57	0.715 47	0.299 51	0.376 55	0.328 58	0.319 52	0.944 27	0.285 53	0.164 44	0.216 59	0.229 55	0.484 50	0.545 53	0.456 51	0.755 48	0.709 49	0.475 52
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
SPLAT Net	0.393 58	0.472 58	0.511 58	0.606 46	0.311 58	0.656 52	0.245 56	0.405 53	0.328 58	0.197 60	0.927 45	0.227 59	0.000 63	0.001 63	0.249 54	0.271 61	0.510 56	0.383 58	0.593 57	0.699 50	0.267 59
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 62	0.203 62	0.366 61	0.501 53	0.311 58	0.524 60	0.211 60	0.002 63	0.342 57	0.189 61	0.786 60	0.145 62	0.102 56	0.245 55	0.152 60	0.318 60	0.348 61	0.300 61	0.460 61	0.437 62	0.182 62
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
subcloud_weak	0.411 57	0.479 57	0.650 49	0.475 55	0.285 60	0.519 61	0.087 62	0.725 25	0.396 54	0.386 48	0.621 62	0.250 56	0.117 53	0.338 52	0.443 45	0.188 62	0.594 49	0.369 59	0.377 62	0.616 57	0.306 57

PointNet++	0.339 60	0.584 49	0.478 60	0.458 57	0.256 61	0.360 62	0.250 54	0.247 59	0.278 61	0.261 58	0.677 61	0.183 60	0.117 53	0.212 60	0.145 61	0.364 57	0.346 62	0.232 62	0.548 58	0.523 61	0.252 60
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 61	0.353 60	0.290 62	0.278 62	0.166 62	0.553 59	0.169 61	0.286 57	0.147 62	0.148 62	0.908 53	0.182 61	0.064 58	0.023 62	0.018 63	0.354 58	0.363 60	0.345 60	0.546 60	0.685 51	0.278 58

ERROR	0.054 63	0.000 63	0.041 63	0.172 63	0.030 63	0.062 63	0.001 63	0.035 62	0.004 63	0.051 63	0.143 63	0.019 63	0.003 62	0.041 61	0.050 62	0.003 63	0.054 63	0.018 63	0.005 63	0.264 63	0.082 63

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

CRNet	0.688 1	1.000 1	0.802 7	0.783 3	0.644 1	0.804 5	0.182 2	0.874 1	0.511 4	0.501 2	0.592 1	0.536 3	0.587 9	1.000 1	0.637 4	0.730 8	0.683 1	0.994 10	0.523 2

Mask-Group	0.664 3	1.000 1	0.822 3	0.764 6	0.616 2	0.815 2	0.139 6	0.694 7	0.597 2	0.459 5	0.566 3	0.599 1	0.600 6	0.516 19	0.715 2	0.819 2	0.635 3	1.000 1	0.603 1

OccuSeg+instance	0.672 2	1.000 1	0.758 11	0.682 7	0.576 3	0.842 1	0.477 1	0.504 15	0.524 3	0.567 1	0.585 2	0.451 5	0.557 10	1.000 1	0.751 1	0.797 3	0.563 8	1.000 1	0.467 7

PE	0.645 5	1.000 1	0.773 8	0.798 2	0.538 4	0.786 9	0.088 12	0.799 4	0.350 10	0.435 9	0.547 5	0.545 2	0.646 5	0.933 7	0.562 7	0.761 5	0.556 12	0.997 8	0.501 5
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
3D-MPA	0.611 8	1.000 1	0.833 2	0.765 5	0.526 5	0.756 10	0.136 8	0.588 12	0.470 5	0.438 8	0.432 12	0.358 11	0.650 2	0.857 9	0.429 11	0.765 4	0.557 10	1.000 1	0.430 9
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 4	1.000 1	0.810 4	0.768 4	0.523 6	0.813 3	0.143 5	0.819 2	0.389 7	0.422 10	0.511 6	0.443 6	0.650 2	1.000 1	0.624 5	0.732 7	0.634 4	1.000 1	0.375 12

Occipital-SCS	0.512 13	1.000 1	0.716 13	0.509 13	0.506 7	0.611 14	0.092 11	0.602 11	0.177 16	0.346 15	0.383 15	0.165 16	0.442 14	0.850 12	0.386 15	0.618 16	0.543 13	0.889 16	0.389 11

PointGroup	0.636 7	1.000 1	0.765 9	0.624 8	0.505 8	0.797 6	0.116 9	0.696 6	0.384 8	0.441 7	0.559 4	0.476 4	0.596 7	1.000 1	0.666 3	0.756 6	0.556 11	0.997 8	0.513 3
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 6	1.000 1	0.895 1	0.800 1	0.480 9	0.676 11	0.144 4	0.737 5	0.354 9	0.447 6	0.400 14	0.365 10	0.700 1	1.000 1	0.569 6	0.836 1	0.599 5	1.000 1	0.473 6

SSEN	0.575 10	1.000 1	0.761 10	0.473 14	0.477 10	0.795 7	0.066 13	0.529 13	0.658 1	0.460 4	0.461 9	0.380 9	0.331 18	0.859 8	0.401 14	0.692 12	0.653 2	1.000 1	0.348 14
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 12	1.000 1	0.538 20	0.282 17	0.468 11	0.790 8	0.173 3	0.345 17	0.429 6	0.413 13	0.484 7	0.176 15	0.595 8	0.591 17	0.522 8	0.668 14	0.476 15	0.986 11	0.327 15

PCJC	0.578 9	1.000 1	0.810 5	0.583 12	0.449 12	0.813 4	0.042 15	0.603 10	0.341 11	0.490 3	0.465 8	0.410 8	0.650 2	0.835 13	0.264 18	0.694 11	0.561 9	0.889 16	0.504 4

MASC	0.447 17	0.528 22	0.555 18	0.381 15	0.382 13	0.633 13	0.002 21	0.509 14	0.260 13	0.361 14	0.432 11	0.327 12	0.451 12	0.571 18	0.367 16	0.639 15	0.386 16	0.980 12	0.276 17
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
R-PointNet	0.306 21	0.500 23	0.405 23	0.311 16	0.348 14	0.589 15	0.054 14	0.068 23	0.126 18	0.283 18	0.290 17	0.028 23	0.219 20	0.214 22	0.331 17	0.396 23	0.275 20	0.821 21	0.245 18

MTML	0.549 11	1.000 1	0.807 6	0.588 11	0.327 15	0.647 12	0.004 20	0.815 3	0.180 15	0.418 11	0.364 16	0.182 14	0.445 13	1.000 1	0.442 10	0.688 13	0.571 7	1.000 1	0.396 10
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
3D-BoNet	0.488 14	1.000 1	0.672 16	0.590 10	0.301 16	0.484 21	0.098 10	0.620 8	0.306 12	0.341 16	0.259 18	0.125 18	0.434 15	0.796 14	0.402 13	0.499 21	0.513 14	0.909 15	0.439 8
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 16	1.000 1	0.737 12	0.159 23	0.259 17	0.587 16	0.138 7	0.475 16	0.217 14	0.416 12	0.408 13	0.128 17	0.315 19	0.714 15	0.411 12	0.536 20	0.590 6	0.873 19	0.304 16
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 15	0.667 17	0.712 15	0.595 9	0.259 18	0.550 19	0.000 23	0.613 9	0.175 17	0.250 20	0.434 10	0.437 7	0.411 17	0.857 9	0.485 9	0.591 19	0.267 22	0.944 13	0.359 13
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 18	1.000 1	0.432 22	0.245 19	0.190 19	0.577 17	0.013 18	0.263 18	0.033 22	0.320 17	0.240 19	0.075 20	0.422 16	0.857 9	0.117 21	0.699 9	0.271 21	0.883 18	0.235 19
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
UNet-backbone	0.319 20	0.667 17	0.715 14	0.233 20	0.189 20	0.479 22	0.008 19	0.218 19	0.067 21	0.201 21	0.173 21	0.107 19	0.123 22	0.438 20	0.150 19	0.615 17	0.355 17	0.916 14	0.093 24

Hier3D	0.323 19	0.667 17	0.542 19	0.264 18	0.157 21	0.550 18	0.000 23	0.205 21	0.009 23	0.270 19	0.218 20	0.075 20	0.500 11	0.688 16	0.007 25	0.698 10	0.301 19	0.459 23	0.200 20
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
Region	0.248 22	0.667 17	0.437 21	0.188 21	0.153 22	0.491 20	0.000 23	0.208 20	0.094 20	0.153 22	0.099 22	0.057 22	0.217 21	0.119 23	0.039 23	0.466 22	0.302 18	0.640 22	0.140 22

Sgpn_scannet	0.143 24	0.208 25	0.390 24	0.169 22	0.065 23	0.275 24	0.029 16	0.069 22	0.000 24	0.087 24	0.043 23	0.014 25	0.027 25	0.000 24	0.112 22	0.351 24	0.168 24	0.438 24	0.138 23

MaskRCNN 2d->3d Proj	0.058 25	0.333 24	0.002 25	0.000 25	0.053 24	0.002 25	0.002 22	0.021 25	0.000 24	0.045 25	0.024 25	0.238 13	0.065 24	0.000 24	0.014 24	0.107 25	0.020 25	0.110 25	0.006 25

3D-BEVIS	0.248 22	0.667 17	0.566 17	0.076 24	0.035 25	0.394 23	0.027 17	0.035 24	0.098 19	0.099 23	0.030 24	0.025 24	0.098 23	0.375 21	0.126 20	0.604 18	0.181 23	0.854 20	0.171 21
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

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 2	0.000 3	0.812 2	0.941 1	0.500 1	0.500 2	0.500 2	0.500 1	0.429 3	0.500 2	0.667 2	0.500 1	0.625 2	0.000 1
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 1	1.000 1	1.000 1	0.500 1	1.000 1	1.000 1	0.778 1	0.000 2	0.938 1	0.000 1
Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler: Indoor Scene Recognition in 3D. IROS 2020
resnet50_scannet	0.353 3	0.250 2	0.812 2	0.529 3	0.500 1	0.500 2	0.000 3	0.500 1	0.571 2	0.000 3	0.556 3	0.000 2	0.375 3	0.000 1

3D Semantic label benchmark

3D Semantic instance benchmark

2D Semantic label benchmark

2D Semantic instance benchmark

Scene type classification benchmark