3D Semantic Instance Results - ScanNet Benchmark

The 3D semantic instance prediction task involves detecting and segmenting the object in an 3D scan mesh.

Evaluation and metrics

Our evaluation ranks all methods according to the average precision for each class. We report the mean average precision AP at overlap 0.25 (AP 25%), overlap 0.5 (AP 50%), and over overlaps in the range [0.5:0.95:0.05] (AP). Note that multiple predictions of the same ground truth instance are penalized as false positives.

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

ExtMask3D	0.789 13	1.000 1	0.988 1	0.756 35	0.706 12	0.912 12	0.429 21	0.647 41	0.806 5	0.755 6	0.673 16	0.689 21	0.772 11	1.000 1	0.789 16	0.852 11	0.811 6	1.000 1	0.617 14

Competitor-SPFormer	0.800 7	1.000 1	0.986 2	0.845 16	0.705 13	0.915 11	0.532 14	0.733 20	0.757 10	0.733 11	0.708 10	0.698 19	0.648 36	0.981 39	0.890 1	0.830 19	0.796 9	0.997 40	0.644 5

Competitor-MAFT	0.816 1	1.000 1	0.983 3	0.872 10	0.718 5	0.941 1	0.588 4	0.652 39	0.819 2	0.776 3	0.720 5	0.780 5	0.769 12	1.000 1	0.797 11	0.813 28	0.798 8	1.000 1	0.659 4

KmaxOneFormerNet	0.783 16	0.903 56	0.981 4	0.794 27	0.706 11	0.931 4	0.561 10	0.701 26	0.706 16	0.727 15	0.697 14	0.731 10	0.689 23	1.000 1	0.856 2	0.750 40	0.761 23	1.000 1	0.599 22

MG-Former	0.791 12	1.000 1	0.980 5	0.837 19	0.626 27	0.897 13	0.543 13	0.759 15	0.800 6	0.766 5	0.659 18	0.769 6	0.697 19	1.000 1	0.791 15	0.707 49	0.791 11	1.000 1	0.610 17

OneFormer3D	0.801 6	1.000 1	0.973 6	0.909 5	0.698 15	0.928 5	0.582 5	0.668 35	0.685 19	0.780 2	0.687 15	0.698 20	0.702 15	1.000 1	0.794 13	0.900 2	0.784 14	0.986 53	0.635 9
Maxim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: OneFormer3D: One Transformer for Unified Point Cloud Segmentation.
Spherical Mask(CtoF)	0.812 3	1.000 1	0.973 7	0.852 14	0.718 6	0.917 10	0.574 6	0.677 30	0.748 11	0.729 14	0.715 8	0.795 2	0.809 1	1.000 1	0.831 4	0.854 9	0.787 12	1.000 1	0.638 7

PointRel	0.816 1	1.000 1	0.971 8	0.908 6	0.743 2	0.923 8	0.573 8	0.714 22	0.695 18	0.734 10	0.747 2	0.725 12	0.809 1	1.000 1	0.814 9	0.899 3	0.820 4	1.000 1	0.610 18
: Relation3D (PointRel): Enhancing Relation Modeling for Point Cloud Instance Segmentation.
EV3D	0.811 4	1.000 1	0.968 9	0.852 14	0.717 7	0.921 9	0.574 7	0.677 30	0.748 11	0.730 13	0.703 13	0.795 2	0.809 1	1.000 1	0.831 4	0.854 9	0.778 16	1.000 1	0.638 8

SIM3D	0.803 5	1.000 1	0.967 10	0.863 13	0.692 19	0.924 7	0.552 12	0.732 21	0.667 22	0.732 12	0.662 17	0.796 1	0.789 9	1.000 1	0.803 10	0.864 6	0.766 21	1.000 1	0.643 6

Queryformer	0.787 14	1.000 1	0.933 11	0.601 51	0.754 1	0.886 18	0.558 11	0.661 37	0.767 9	0.665 20	0.716 7	0.639 26	0.808 5	1.000 1	0.844 3	0.897 4	0.804 7	1.000 1	0.624 12

UniPerception	0.800 7	1.000 1	0.930 12	0.872 10	0.727 4	0.862 25	0.454 20	0.764 13	0.820 1	0.746 7	0.706 11	0.750 7	0.772 10	0.926 46	0.764 19	0.818 27	0.826 2	0.997 40	0.660 3

TST3D	0.795 11	1.000 1	0.929 13	0.918 4	0.709 10	0.884 20	0.596 3	0.704 25	0.769 8	0.734 9	0.644 22	0.699 18	0.751 13	1.000 1	0.794 12	0.876 5	0.757 24	0.997 40	0.550 33
Duc Tran Dang Trung, Byeongkeun Kang, Yeejin Lee: MSTA3D: Multi-scale Twin-attention for 3D Instance Segmentation. ACM Multimedia 2024
InsSSM	0.799 9	1.000 1	0.915 14	0.710 42	0.729 3	0.925 6	0.664 1	0.670 33	0.770 7	0.766 4	0.739 3	0.737 8	0.700 16	1.000 1	0.792 14	0.829 21	0.815 5	0.997 40	0.625 11
Lei Yao, Yi Wang, Moyun Liu, Lap-Pui Chau: SGIFormer: Semantic-guided and Geometric-enhanced Interleaving Transformer for 3D Instance Segmentation. TCSVT, 2024
ISBNet	0.757 21	1.000 1	0.904 15	0.731 38	0.678 21	0.895 14	0.458 18	0.644 43	0.670 21	0.710 17	0.620 27	0.732 9	0.650 26	1.000 1	0.756 20	0.778 32	0.779 15	1.000 1	0.614 15
Tuan Duc Ngo, Binh-Son Hua, Khoi Nguyen: ISBNet: a 3D Point Cloud Instance Segmentation Network with Instance-aware Sampling and Box-aware Dynamic Convolution. CVPR 2023
SPFormer	0.770 18	0.903 56	0.903 16	0.806 24	0.609 33	0.886 17	0.568 9	0.815 6	0.705 17	0.711 16	0.655 19	0.652 25	0.685 24	1.000 1	0.789 17	0.809 29	0.776 18	1.000 1	0.583 26
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
NeuralBF	0.555 53	0.667 59	0.896 17	0.843 18	0.517 48	0.751 49	0.029 58	0.519 56	0.414 45	0.439 45	0.465 47	0.000 75	0.484 54	0.857 48	0.287 63	0.693 55	0.651 38	1.000 1	0.485 44
Weiwei Sun, Daniel Rebain, Renjie Liao, Vladimir Tankovich, Soroosh Yazdani, Kwang Moo Yi, Andrea Tagliasacchi: NeuralBF: Neural Bilateral Filtering for Top-down Instance Segmentation on Point Clouds. WACV 2023
GICN	0.638 42	1.000 1	0.895 18	0.800 25	0.480 53	0.676 55	0.144 41	0.737 18	0.354 50	0.447 43	0.400 57	0.365 52	0.700 16	1.000 1	0.569 41	0.836 17	0.599 43	1.000 1	0.473 45

MAFT	0.786 15	1.000 1	0.894 19	0.807 23	0.694 17	0.893 16	0.486 16	0.674 32	0.740 13	0.786 1	0.704 12	0.727 11	0.739 14	1.000 1	0.707 25	0.849 13	0.756 25	1.000 1	0.685 2

DCD	0.798 10	1.000 1	0.878 20	0.792 28	0.693 18	0.936 2	0.596 2	0.685 29	0.663 24	0.736 8	0.717 6	0.788 4	0.693 21	1.000 1	0.825 7	0.840 15	0.837 1	1.000 1	0.689 1

SphereSeg	0.680 32	1.000 1	0.856 21	0.744 36	0.618 31	0.893 15	0.151 39	0.651 40	0.713 14	0.537 31	0.579 36	0.430 45	0.651 25	1.000 1	0.389 59	0.744 43	0.697 29	0.991 51	0.601 21

SSEC	0.707 28	1.000 1	0.850 22	0.924 3	0.648 24	0.747 50	0.162 38	0.862 3	0.572 32	0.520 32	0.624 26	0.549 31	0.649 35	1.000 1	0.560 43	0.706 50	0.768 20	1.000 1	0.591 25

HAIS	0.699 29	1.000 1	0.849 23	0.820 21	0.675 22	0.808 37	0.279 28	0.757 16	0.465 39	0.517 33	0.596 31	0.559 30	0.600 42	1.000 1	0.654 31	0.767 35	0.676 32	0.994 49	0.560 31
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
Box2Mask	0.677 34	1.000 1	0.847 24	0.771 31	0.509 49	0.816 33	0.277 29	0.558 52	0.482 36	0.562 28	0.640 23	0.448 41	0.700 16	1.000 1	0.666 27	0.852 12	0.578 48	0.997 40	0.488 43
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
Dyco3D	0.641 41	1.000 1	0.841 25	0.893 7	0.531 44	0.802 39	0.115 48	0.588 50	0.448 41	0.438 46	0.537 43	0.430 46	0.550 51	0.857 48	0.534 46	0.764 37	0.657 34	0.987 52	0.568 28
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
3D-MPA	0.611 47	1.000 1	0.833 26	0.765 33	0.526 46	0.756 48	0.136 45	0.588 50	0.470 38	0.438 47	0.432 53	0.358 54	0.650 26	0.857 48	0.429 55	0.765 36	0.557 54	1.000 1	0.430 50
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
Mask3D_evaluation	0.631 45	1.000 1	0.829 27	0.606 50	0.646 25	0.836 29	0.068 52	0.511 57	0.462 40	0.507 35	0.619 28	0.389 50	0.610 41	1.000 1	0.432 54	0.828 22	0.673 33	0.788 68	0.552 32

Mask-Group	0.664 36	1.000 1	0.822 28	0.764 34	0.616 32	0.815 34	0.139 43	0.694 28	0.597 30	0.459 42	0.566 38	0.599 29	0.600 42	0.516 66	0.715 24	0.819 26	0.635 39	1.000 1	0.603 19
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
PBNet	0.747 23	1.000 1	0.818 29	0.837 20	0.713 9	0.844 27	0.457 19	0.647 41	0.711 15	0.614 22	0.617 29	0.657 24	0.650 26	1.000 1	0.692 26	0.822 23	0.765 22	1.000 1	0.595 23
Weiguang Zhao, Yuyao Yan, Chaolong Yang, Jianan Ye, Xi Yang, Kaizhu Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
DKNet	0.718 27	1.000 1	0.814 30	0.782 29	0.619 30	0.872 21	0.224 34	0.751 17	0.569 33	0.677 18	0.585 33	0.724 13	0.633 38	0.981 39	0.515 48	0.819 25	0.736 28	1.000 1	0.617 13
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
CSC-Pretrained	0.648 38	1.000 1	0.810 31	0.768 32	0.523 47	0.813 35	0.143 42	0.819 5	0.389 47	0.422 51	0.511 44	0.443 42	0.650 26	1.000 1	0.624 36	0.732 45	0.634 40	1.000 1	0.375 55

PCJC	0.578 50	1.000 1	0.810 32	0.583 56	0.449 56	0.813 36	0.042 57	0.603 48	0.341 52	0.490 37	0.465 48	0.410 47	0.650 26	0.835 56	0.264 65	0.694 54	0.561 52	0.889 61	0.504 40

SoftGroup	0.761 20	1.000 1	0.808 33	0.845 16	0.716 8	0.862 24	0.243 33	0.824 4	0.655 26	0.620 21	0.734 4	0.699 17	0.791 8	0.981 39	0.716 23	0.844 14	0.769 19	1.000 1	0.594 24
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
DANCENET	0.680 32	1.000 1	0.807 34	0.733 37	0.600 34	0.768 46	0.375 25	0.543 53	0.538 34	0.610 23	0.599 30	0.498 36	0.632 40	0.981 39	0.739 22	0.856 8	0.633 41	0.882 64	0.454 48

MTML	0.549 54	1.000 1	0.807 35	0.588 55	0.327 61	0.647 57	0.004 65	0.815 7	0.180 60	0.418 52	0.364 61	0.182 59	0.445 57	1.000 1	0.442 53	0.688 57	0.571 50	1.000 1	0.396 53
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
TopoSeg	0.725 26	1.000 1	0.806 36	0.933 2	0.668 23	0.758 47	0.272 30	0.734 19	0.630 28	0.549 30	0.654 20	0.606 28	0.697 20	0.966 43	0.612 38	0.839 16	0.754 26	1.000 1	0.573 27

SoftGroup++	0.769 19	1.000 1	0.803 37	0.937 1	0.684 20	0.865 22	0.213 36	0.870 2	0.664 23	0.571 26	0.758 1	0.702 16	0.807 6	1.000 1	0.653 32	0.902 1	0.792 10	1.000 1	0.626 10

OSIS	0.605 48	1.000 1	0.801 38	0.599 52	0.535 43	0.728 52	0.286 27	0.436 63	0.679 20	0.491 36	0.433 51	0.256 56	0.404 63	0.857 48	0.620 37	0.724 46	0.510 61	1.000 1	0.539 34

DualGroup	0.694 31	1.000 1	0.799 39	0.811 22	0.622 28	0.817 32	0.376 24	0.805 9	0.590 31	0.487 38	0.568 37	0.525 35	0.650 26	0.835 56	0.600 39	0.829 20	0.655 35	1.000 1	0.526 35

DENet	0.629 46	1.000 1	0.797 40	0.608 49	0.589 35	0.627 59	0.219 35	0.882 1	0.310 53	0.402 56	0.383 59	0.396 49	0.650 26	1.000 1	0.663 29	0.543 67	0.691 31	1.000 1	0.568 29

DD-UNet+Group	0.635 44	0.667 59	0.797 41	0.714 41	0.562 39	0.774 45	0.146 40	0.810 8	0.429 43	0.476 39	0.546 42	0.399 48	0.633 38	1.000 1	0.632 35	0.722 47	0.609 42	1.000 1	0.514 36
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
GraphCut	0.732 24	1.000 1	0.788 42	0.724 39	0.642 26	0.859 26	0.248 32	0.787 11	0.618 29	0.596 25	0.653 21	0.722 14	0.583 48	1.000 1	0.766 18	0.861 7	0.825 3	1.000 1	0.504 39

IPCA-Inst	0.731 25	1.000 1	0.788 43	0.884 9	0.698 14	0.788 43	0.252 31	0.760 14	0.646 27	0.511 34	0.637 24	0.665 23	0.804 7	1.000 1	0.644 33	0.778 33	0.747 27	1.000 1	0.561 30

Mask3D	0.780 17	1.000 1	0.786 44	0.716 40	0.696 16	0.885 19	0.500 15	0.714 22	0.810 4	0.672 19	0.715 8	0.679 22	0.809 1	1.000 1	0.831 4	0.833 18	0.787 12	1.000 1	0.602 20
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
TD3D	0.751 22	1.000 1	0.774 45	0.867 12	0.621 29	0.934 3	0.404 22	0.706 24	0.812 3	0.605 24	0.633 25	0.626 27	0.690 22	1.000 1	0.640 34	0.820 24	0.777 17	1.000 1	0.612 16
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
SegGroup_ins	0.445 64	0.667 59	0.773 46	0.185 70	0.317 62	0.656 56	0.000 68	0.407 64	0.134 64	0.381 59	0.267 64	0.217 58	0.476 55	0.714 59	0.452 52	0.629 62	0.514 59	1.000 1	0.222 66
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
PE	0.645 39	1.000 1	0.773 47	0.798 26	0.538 42	0.786 44	0.088 51	0.799 10	0.350 51	0.435 49	0.547 41	0.545 32	0.646 37	0.933 45	0.562 42	0.761 38	0.556 56	0.997 40	0.501 41
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
PointGroup	0.636 43	1.000 1	0.765 48	0.624 48	0.505 51	0.797 40	0.116 47	0.696 27	0.384 48	0.441 44	0.559 39	0.476 38	0.596 45	1.000 1	0.666 27	0.756 39	0.556 55	0.997 40	0.513 37
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]
SSEN	0.575 51	1.000 1	0.761 49	0.473 59	0.477 54	0.795 41	0.066 53	0.529 55	0.658 25	0.460 41	0.461 49	0.380 51	0.331 65	0.859 47	0.401 58	0.692 56	0.653 36	1.000 1	0.348 57
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
AOIA	0.601 49	1.000 1	0.761 50	0.687 43	0.485 52	0.828 30	0.008 63	0.663 36	0.405 46	0.405 55	0.425 54	0.490 37	0.596 45	0.714 59	0.553 45	0.779 31	0.597 44	0.992 50	0.424 52

INS-Conv-instance	0.657 37	1.000 1	0.760 51	0.667 46	0.581 36	0.863 23	0.323 26	0.655 38	0.477 37	0.473 40	0.549 40	0.432 44	0.650 26	1.000 1	0.655 30	0.738 44	0.585 47	0.944 56	0.472 46

RPGN	0.643 40	1.000 1	0.758 52	0.582 57	0.539 41	0.826 31	0.046 56	0.765 12	0.372 49	0.436 48	0.588 32	0.539 34	0.650 26	1.000 1	0.577 40	0.750 41	0.653 37	0.997 40	0.495 42
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
OccuSeg+instance	0.672 35	1.000 1	0.758 53	0.682 44	0.576 38	0.842 28	0.477 17	0.504 59	0.524 35	0.567 27	0.585 35	0.451 40	0.557 50	1.000 1	0.751 21	0.797 30	0.563 51	1.000 1	0.467 47
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
Region-18class	0.284 69	0.250 75	0.751 54	0.228 68	0.270 64	0.521 67	0.000 68	0.468 62	0.008 74	0.205 68	0.127 69	0.000 75	0.068 73	0.070 74	0.262 66	0.652 60	0.323 66	0.740 69	0.173 68

SALoss-ResNet	0.459 62	1.000 1	0.737 55	0.159 73	0.259 65	0.587 62	0.138 44	0.475 61	0.217 59	0.416 53	0.408 55	0.128 62	0.315 66	0.714 59	0.411 56	0.536 68	0.590 46	0.873 65	0.304 60
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)
One_Thing_One_Click	0.529 56	0.667 59	0.718 56	0.777 30	0.399 57	0.683 54	0.000 68	0.669 34	0.138 63	0.391 58	0.374 60	0.539 33	0.360 64	0.641 63	0.556 44	0.774 34	0.593 45	0.997 40	0.251 63
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
Occipital-SCS	0.512 58	1.000 1	0.716 57	0.509 58	0.506 50	0.611 60	0.092 50	0.602 49	0.177 61	0.346 61	0.383 58	0.165 61	0.442 58	0.850 55	0.386 60	0.618 63	0.543 58	0.889 61	0.389 54

UNet-backbone	0.319 67	0.667 59	0.715 58	0.233 66	0.189 68	0.479 71	0.008 63	0.218 68	0.067 69	0.201 69	0.173 68	0.107 64	0.123 71	0.438 67	0.150 67	0.615 64	0.355 64	0.916 59	0.093 75

PanopticFusion-inst	0.478 60	0.667 59	0.712 59	0.595 53	0.259 66	0.550 66	0.000 68	0.613 47	0.175 62	0.250 67	0.434 50	0.437 43	0.411 62	0.857 48	0.485 49	0.591 66	0.267 72	0.944 56	0.359 56
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
RWSeg	0.567 52	0.528 69	0.708 60	0.626 47	0.580 37	0.745 51	0.063 54	0.627 44	0.240 57	0.400 57	0.497 45	0.464 39	0.515 52	1.000 1	0.475 50	0.745 42	0.571 49	1.000 1	0.429 51

SSTNet	0.698 30	1.000 1	0.697 61	0.888 8	0.556 40	0.803 38	0.387 23	0.626 45	0.417 44	0.556 29	0.585 34	0.702 15	0.600 42	1.000 1	0.824 8	0.720 48	0.692 30	1.000 1	0.509 38
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
SPG_WSIS	0.470 61	0.667 59	0.685 62	0.677 45	0.372 59	0.562 64	0.000 68	0.482 60	0.244 56	0.316 64	0.298 62	0.052 70	0.442 59	0.857 48	0.267 64	0.702 51	0.559 53	1.000 1	0.287 61

3D-BoNet	0.488 59	1.000 1	0.672 63	0.590 54	0.301 63	0.484 70	0.098 49	0.620 46	0.306 54	0.341 62	0.259 65	0.125 63	0.434 60	0.796 58	0.402 57	0.499 69	0.513 60	0.909 60	0.439 49
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
ClickSeg_Instance	0.539 55	1.000 1	0.621 64	0.300 62	0.530 45	0.698 53	0.127 46	0.533 54	0.222 58	0.430 50	0.400 56	0.365 52	0.574 49	0.938 44	0.472 51	0.659 59	0.543 57	0.944 56	0.347 58

SemRegionNet-20cls	0.250 70	0.333 72	0.613 65	0.229 67	0.163 69	0.493 68	0.000 68	0.304 66	0.107 66	0.147 72	0.100 71	0.052 69	0.231 67	0.119 72	0.039 72	0.445 71	0.325 65	0.654 70	0.141 71

3D-BEVIS	0.248 71	0.667 59	0.566 66	0.076 74	0.035 76	0.394 74	0.027 60	0.035 75	0.098 67	0.099 74	0.030 75	0.025 72	0.098 72	0.375 69	0.126 69	0.604 65	0.181 74	0.854 66	0.171 69
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.
MASC	0.447 63	0.528 69	0.555 67	0.381 60	0.382 58	0.633 58	0.002 66	0.509 58	0.260 55	0.361 60	0.432 52	0.327 55	0.451 56	0.571 65	0.367 61	0.639 61	0.386 63	0.980 55	0.276 62
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
Hier3D	0.323 66	0.667 59	0.542 68	0.264 64	0.157 70	0.550 65	0.000 68	0.205 70	0.009 72	0.270 66	0.218 67	0.075 66	0.500 53	0.688 62	0.007 76	0.698 53	0.301 68	0.459 73	0.200 67
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
Sparse R-CNN	0.515 57	1.000 1	0.538 69	0.282 63	0.468 55	0.790 42	0.173 37	0.345 65	0.429 42	0.413 54	0.484 46	0.176 60	0.595 47	0.591 64	0.522 47	0.668 58	0.476 62	0.986 54	0.327 59

Sem_Recon_ins	0.227 73	0.764 58	0.486 70	0.069 75	0.098 73	0.426 73	0.017 61	0.067 74	0.015 71	0.172 70	0.100 70	0.096 65	0.054 75	0.183 71	0.135 68	0.366 74	0.260 73	0.614 72	0.168 70

tmp	0.248 71	0.667 59	0.437 71	0.188 69	0.153 71	0.491 69	0.000 68	0.208 69	0.094 68	0.153 71	0.099 72	0.057 68	0.217 70	0.119 72	0.039 72	0.466 70	0.302 67	0.640 71	0.140 72

3D-SIS	0.382 65	1.000 1	0.432 72	0.245 65	0.190 67	0.577 63	0.013 62	0.263 67	0.033 70	0.320 63	0.240 66	0.075 66	0.422 61	0.857 48	0.117 70	0.699 52	0.271 71	0.883 63	0.235 65
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
R-PointNet	0.306 68	0.500 71	0.405 73	0.311 61	0.348 60	0.589 61	0.054 55	0.068 73	0.126 65	0.283 65	0.290 63	0.028 71	0.219 69	0.214 70	0.331 62	0.396 73	0.275 69	0.821 67	0.245 64

Sgpn_scannet	0.143 75	0.208 76	0.390 74	0.169 71	0.065 74	0.275 75	0.029 59	0.069 72	0.000 75	0.087 75	0.043 74	0.014 73	0.027 76	0.000 75	0.112 71	0.351 75	0.168 75	0.438 74	0.138 73

ASIS	0.199 74	0.333 72	0.253 75	0.167 72	0.140 72	0.438 72	0.000 68	0.177 71	0.008 73	0.121 73	0.069 73	0.004 74	0.231 68	0.429 68	0.036 74	0.445 72	0.273 70	0.333 75	0.119 74

MaskRCNN 2d->3d Proj	0.058 76	0.333 72	0.002 76	0.000 76	0.053 75	0.002 76	0.002 67	0.021 76	0.000 75	0.045 76	0.024 76	0.238 57	0.065 74	0.000 75	0.014 75	0.107 76	0.020 76	0.110 76	0.006 76

3D Semantic Instance Benchmark