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

GraphCut	0.552 20	1.000 1	0.611 19	0.438 27	0.392 22	0.714 23	0.139 24	0.598 13	0.327 22	0.389 23	0.510 17	0.598 12	0.427 39	0.754 21	0.463 18	0.761 3	0.588 9	0.903 23	0.329 29

InsSSM	0.586 11	1.000 1	0.593 21	0.440 26	0.480 5	0.771 8	0.345 1	0.437 40	0.444 7	0.495 14	0.548 5	0.579 16	0.621 11	0.720 28	0.409 23	0.712 11	0.593 6	0.960 5	0.395 10
Lei Yao, Yi Wang, Moyun Liu, Lap-Pui Chau: SGIFormer: Semantic-guided and Geometric-enhanced Interleaving Transformer for 3D Instance Segmentation. TCSVT, 2024
UniPerception	0.588 9	0.963 3	0.667 10	0.493 15	0.472 7	0.750 15	0.229 19	0.528 27	0.468 4	0.498 13	0.542 8	0.643 8	0.530 21	0.661 38	0.463 16	0.695 18	0.599 3	0.972 1	0.420 7

SIM3D	0.617 3	0.952 4	0.629 17	0.539 11	0.426 16	0.768 11	0.302 6	0.681 2	0.425 9	0.473 16	0.511 16	0.701 2	0.717 1	0.821 6	0.467 15	0.774 1	0.559 15	0.914 18	0.448 3

EV3D	0.615 5	0.946 5	0.652 13	0.555 6	0.433 14	0.773 6	0.271 13	0.604 8	0.447 5	0.506 7	0.544 7	0.698 3	0.716 2	0.775 16	0.480 9	0.747 5	0.572 13	0.925 14	0.435 6

Spherical Mask(CtoF)	0.616 4	0.946 5	0.654 12	0.555 6	0.434 13	0.769 10	0.271 12	0.604 8	0.447 5	0.505 8	0.549 3	0.698 3	0.716 2	0.775 16	0.480 9	0.747 5	0.575 11	0.925 14	0.436 5

ISBNet	0.559 19	0.939 7	0.655 11	0.383 40	0.426 17	0.763 13	0.180 21	0.534 25	0.386 13	0.499 11	0.509 18	0.621 10	0.427 38	0.704 33	0.467 14	0.649 23	0.571 14	0.948 10	0.401 9
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
DANCENET	0.504 26	0.926 8	0.579 22	0.472 18	0.367 27	0.626 43	0.165 23	0.432 41	0.221 34	0.408 22	0.449 30	0.411 25	0.564 17	0.746 24	0.421 22	0.707 17	0.438 33	0.846 43	0.288 39

PBNet	0.573 15	0.926 8	0.575 26	0.619 2	0.472 6	0.736 19	0.239 18	0.487 37	0.383 15	0.459 19	0.506 19	0.533 22	0.585 14	0.767 18	0.404 24	0.717 10	0.559 16	0.969 2	0.381 16
Weiguang Zhao, Yuyao Yan, Chaolong Yang, Jianan Ye, Xi Yang, Kaizhu Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
PointRel	0.622 1	0.926 8	0.710 3	0.541 10	0.502 2	0.772 7	0.314 4	0.598 11	0.425 8	0.504 10	0.565 2	0.650 7	0.716 2	0.809 7	0.476 12	0.747 4	0.618 2	0.963 4	0.364 20
: Relation3D (PointRel): Enhancing Relation Modeling for Point Cloud Instance Segmentation.
Queryformer	0.583 12	0.926 8	0.702 5	0.393 37	0.504 1	0.733 21	0.276 11	0.527 28	0.373 17	0.479 15	0.534 11	0.533 23	0.697 7	0.720 29	0.436 21	0.745 7	0.592 7	0.958 7	0.363 21

Mask3D	0.566 17	0.926 8	0.597 20	0.408 34	0.420 18	0.737 18	0.239 17	0.598 11	0.386 14	0.458 20	0.549 3	0.568 19	0.716 2	0.601 44	0.480 9	0.646 25	0.575 11	0.922 16	0.364 19
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
DCD	0.614 6	0.892 13	0.633 16	0.434 28	0.495 3	0.810 1	0.292 8	0.501 33	0.408 10	0.525 4	0.582 1	0.688 5	0.625 10	0.801 8	0.608 1	0.672 20	0.649 1	0.965 3	0.476 1

IPCA-Inst	0.520 23	0.889 14	0.551 30	0.548 9	0.418 19	0.665 33	0.064 35	0.585 14	0.260 32	0.277 37	0.471 25	0.500 24	0.644 9	0.785 11	0.369 25	0.591 34	0.511 22	0.878 32	0.362 23

MAFT	0.596 8	0.889 14	0.721 2	0.448 23	0.460 9	0.768 12	0.251 16	0.558 21	0.408 11	0.504 9	0.539 9	0.616 11	0.618 12	0.858 3	0.482 8	0.684 19	0.551 18	0.931 13	0.450 2

Competitor-MAFT	0.618 2	0.866 16	0.724 1	0.628 1	0.484 4	0.803 2	0.300 7	0.509 32	0.496 1	0.539 1	0.547 6	0.703 1	0.668 8	0.708 31	0.463 17	0.708 16	0.595 4	0.959 6	0.418 8

ExtMask3D	0.598 7	0.852 17	0.692 7	0.433 31	0.461 8	0.791 4	0.264 14	0.488 36	0.493 2	0.508 6	0.528 15	0.594 13	0.706 6	0.791 10	0.483 7	0.734 9	0.595 5	0.911 20	0.437 4

TD3D	0.489 28	0.852 17	0.511 40	0.434 29	0.322 34	0.735 20	0.101 29	0.512 30	0.355 20	0.349 27	0.468 26	0.283 35	0.514 24	0.676 37	0.268 39	0.671 21	0.510 23	0.908 21	0.329 30
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
MG-Former	0.587 10	0.852 17	0.639 15	0.454 22	0.393 21	0.758 14	0.338 2	0.572 16	0.480 3	0.527 3	0.491 22	0.671 6	0.527 22	0.867 1	0.485 6	0.601 31	0.590 8	0.938 12	0.390 12

SSEN	0.384 46	0.852 17	0.494 46	0.192 58	0.226 51	0.648 38	0.022 50	0.398 47	0.299 28	0.277 36	0.317 46	0.231 44	0.194 61	0.514 51	0.196 48	0.586 36	0.444 32	0.843 46	0.184 54
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
DualGroup	0.469 31	0.815 21	0.552 29	0.398 35	0.374 25	0.683 29	0.130 25	0.539 24	0.310 23	0.327 30	0.407 34	0.276 36	0.447 32	0.535 48	0.342 28	0.659 22	0.455 31	0.900 26	0.301 34

DKNet	0.532 22	0.815 21	0.624 18	0.517 12	0.377 24	0.749 16	0.107 26	0.509 31	0.304 24	0.437 21	0.475 23	0.581 15	0.539 19	0.775 15	0.339 29	0.640 27	0.506 24	0.901 24	0.385 15
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
OccuSeg+instance	0.486 29	0.802 23	0.536 33	0.428 32	0.369 26	0.702 25	0.205 20	0.331 54	0.301 26	0.379 25	0.474 24	0.327 28	0.437 33	0.862 2	0.485 5	0.601 32	0.394 41	0.846 45	0.273 42
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
OneFormer3D	0.566 17	0.781 24	0.697 6	0.562 5	0.431 15	0.770 9	0.331 3	0.400 46	0.373 18	0.529 2	0.504 20	0.568 18	0.475 28	0.732 26	0.470 13	0.762 2	0.550 19	0.871 35	0.379 17
Maxim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: OneFormer3D: One Transformer for Unified Point Cloud Segmentation.
Mask-Group	0.434 36	0.778 25	0.516 38	0.471 19	0.330 33	0.658 34	0.029 47	0.526 29	0.249 33	0.256 40	0.400 35	0.309 32	0.384 47	0.296 64	0.368 26	0.575 38	0.425 36	0.877 33	0.362 24
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
OSIS	0.392 44	0.778 25	0.530 34	0.220 57	0.278 41	0.567 51	0.083 31	0.330 55	0.299 27	0.270 38	0.310 47	0.143 54	0.260 54	0.624 42	0.277 37	0.568 41	0.361 46	0.865 37	0.301 33

TST3D	0.569 16	0.778 25	0.675 9	0.598 3	0.451 11	0.727 22	0.280 10	0.476 39	0.395 12	0.472 17	0.457 28	0.583 14	0.580 16	0.777 13	0.462 19	0.735 8	0.547 20	0.919 17	0.333 28
Duc Tran Dang Trung, Byeongkeun Kang, Yeejin Lee: MSTA3D: Multi-scale Twin-attention for 3D Instance Segmentation. ACM Multimedia 2024
KmaxOneFormerNet	0.581 13	0.745 28	0.692 8	0.551 8	0.458 10	0.798 3	0.264 15	0.531 26	0.369 19	0.513 5	0.531 14	0.632 9	0.494 25	0.798 9	0.567 3	0.648 24	0.558 17	0.950 9	0.362 22

SPFormer	0.549 21	0.745 28	0.640 14	0.484 16	0.395 20	0.739 17	0.311 5	0.566 18	0.335 21	0.468 18	0.492 21	0.555 20	0.478 27	0.747 23	0.436 20	0.712 12	0.540 21	0.893 27	0.343 27
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
Box2Mask	0.433 37	0.741 30	0.463 50	0.433 30	0.283 40	0.625 44	0.103 27	0.298 59	0.125 49	0.260 39	0.424 33	0.322 30	0.472 29	0.701 34	0.363 27	0.711 13	0.309 58	0.882 30	0.272 44
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
DENet	0.413 39	0.741 30	0.520 35	0.237 55	0.284 39	0.523 54	0.097 30	0.691 1	0.138 44	0.209 54	0.229 56	0.238 42	0.390 45	0.707 32	0.310 31	0.448 59	0.470 29	0.892 28	0.310 32

SSTNet	0.506 25	0.738 32	0.549 31	0.497 14	0.316 35	0.693 27	0.178 22	0.377 49	0.198 38	0.330 28	0.463 27	0.576 17	0.515 23	0.857 4	0.494 4	0.637 28	0.457 30	0.943 11	0.290 38
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
CSC-Pretrained	0.405 41	0.738 32	0.465 49	0.331 48	0.205 53	0.655 36	0.051 40	0.601 10	0.092 54	0.211 53	0.329 44	0.198 47	0.459 30	0.775 14	0.195 50	0.524 50	0.400 40	0.878 31	0.184 55

Competitor-SPFormer	0.580 14	0.721 34	0.705 4	0.593 4	0.444 12	0.786 5	0.286 9	0.564 19	0.376 16	0.498 12	0.534 12	0.546 21	0.390 44	0.785 12	0.577 2	0.708 15	0.579 10	0.954 8	0.388 13

INS-Conv-instance	0.435 35	0.716 35	0.495 45	0.355 44	0.331 32	0.689 28	0.102 28	0.394 48	0.208 37	0.280 35	0.395 36	0.250 39	0.544 18	0.741 25	0.309 32	0.536 48	0.391 42	0.842 48	0.258 46

Sparse R-CNN	0.292 57	0.704 36	0.213 67	0.153 61	0.154 58	0.551 53	0.053 38	0.212 65	0.132 46	0.174 59	0.274 51	0.070 60	0.363 48	0.441 57	0.176 53	0.424 61	0.234 62	0.758 54	0.161 59

SoftGroup++	0.513 24	0.704 36	0.578 25	0.398 36	0.363 30	0.704 24	0.061 36	0.647 5	0.297 29	0.378 26	0.537 10	0.343 27	0.614 13	0.828 5	0.295 34	0.710 14	0.505 26	0.875 34	0.394 11

TopoSeg	0.479 30	0.704 36	0.564 27	0.467 20	0.366 28	0.633 41	0.068 33	0.554 22	0.262 31	0.328 29	0.447 31	0.323 29	0.534 20	0.722 27	0.288 36	0.614 29	0.482 28	0.912 19	0.358 25

HAIS	0.457 33	0.704 36	0.561 28	0.457 21	0.364 29	0.673 30	0.046 44	0.547 23	0.194 39	0.308 32	0.426 32	0.288 34	0.454 31	0.711 30	0.262 40	0.563 42	0.434 35	0.889 29	0.344 26
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
PCJC	0.375 48	0.704 36	0.542 32	0.284 52	0.197 55	0.649 37	0.006 59	0.426 42	0.138 45	0.242 42	0.304 48	0.183 51	0.388 46	0.629 41	0.141 61	0.546 46	0.344 49	0.738 56	0.283 40

AOIA	0.387 45	0.704 36	0.515 39	0.385 39	0.225 52	0.669 31	0.005 62	0.482 38	0.126 48	0.181 57	0.269 53	0.221 45	0.426 40	0.478 54	0.218 46	0.592 33	0.371 44	0.851 39	0.242 48

Occipital-SCS	0.320 56	0.679 42	0.352 57	0.334 47	0.229 50	0.436 58	0.025 48	0.412 45	0.058 59	0.161 62	0.240 55	0.085 58	0.262 53	0.496 53	0.187 51	0.467 56	0.328 54	0.775 52	0.231 50

SSEC	0.465 32	0.667 43	0.578 24	0.502 13	0.362 31	0.641 40	0.035 45	0.605 7	0.291 30	0.323 31	0.451 29	0.296 33	0.417 42	0.677 36	0.245 43	0.501 52	0.506 25	0.900 25	0.366 18

SALoss-ResNet	0.262 59	0.667 43	0.335 58	0.067 68	0.123 62	0.427 60	0.022 49	0.280 63	0.058 58	0.216 49	0.211 59	0.039 64	0.142 64	0.519 50	0.106 65	0.338 65	0.310 57	0.721 59	0.138 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)
SoftGroup	0.504 26	0.667 43	0.579 23	0.372 42	0.381 23	0.694 26	0.072 32	0.677 3	0.303 25	0.387 24	0.531 13	0.319 31	0.582 15	0.754 20	0.318 30	0.643 26	0.492 27	0.907 22	0.388 14
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
PE	0.396 42	0.667 43	0.467 48	0.446 25	0.243 48	0.624 46	0.022 51	0.577 15	0.106 51	0.219 47	0.340 42	0.239 41	0.487 26	0.475 55	0.225 45	0.541 47	0.350 48	0.818 50	0.273 43
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
ClickSeg_Instance	0.366 49	0.654 47	0.375 54	0.184 59	0.302 38	0.592 49	0.050 42	0.300 58	0.093 53	0.283 33	0.277 50	0.249 40	0.426 41	0.615 43	0.299 33	0.504 51	0.367 45	0.832 49	0.191 53

SphereSeg	0.357 50	0.651 48	0.411 52	0.345 45	0.264 45	0.630 42	0.059 37	0.289 61	0.212 35	0.240 43	0.336 43	0.158 53	0.305 51	0.557 45	0.159 57	0.455 58	0.341 51	0.726 58	0.294 35

Dyco3D	0.395 43	0.642 49	0.518 37	0.447 24	0.259 46	0.666 32	0.050 41	0.251 64	0.166 41	0.231 45	0.362 40	0.232 43	0.331 50	0.535 47	0.229 44	0.587 35	0.438 34	0.850 40	0.317 31
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
PointGroup	0.407 40	0.639 50	0.496 44	0.415 33	0.243 49	0.645 39	0.021 52	0.570 17	0.114 50	0.211 52	0.359 41	0.217 46	0.428 37	0.660 39	0.256 41	0.562 43	0.341 50	0.860 38	0.291 36
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]
RPGN	0.428 38	0.630 51	0.508 42	0.367 43	0.249 47	0.658 35	0.016 55	0.673 4	0.131 47	0.234 44	0.383 39	0.270 37	0.434 34	0.748 22	0.274 38	0.609 30	0.406 38	0.842 47	0.267 45
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
DD-UNet+Group	0.436 34	0.630 51	0.508 43	0.480 17	0.310 37	0.624 45	0.065 34	0.638 6	0.174 40	0.256 41	0.384 38	0.194 48	0.428 36	0.759 19	0.289 35	0.574 39	0.400 39	0.849 42	0.291 37
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
NeuralBF	0.353 52	0.593 53	0.511 41	0.375 41	0.264 44	0.597 48	0.008 57	0.332 53	0.160 42	0.229 46	0.274 52	0.000 75	0.206 58	0.678 35	0.155 58	0.485 54	0.422 37	0.816 51	0.254 47
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
Mask3D_evaluation	0.382 47	0.593 53	0.520 36	0.390 38	0.314 36	0.600 47	0.018 54	0.287 62	0.151 43	0.281 34	0.387 37	0.169 52	0.429 35	0.654 40	0.172 54	0.578 37	0.384 43	0.670 61	0.278 41

GICN	0.341 54	0.580 55	0.371 55	0.344 46	0.198 54	0.469 57	0.052 39	0.564 20	0.093 52	0.212 50	0.212 58	0.127 56	0.347 49	0.537 46	0.206 47	0.525 49	0.329 53	0.729 57	0.241 49

MTML	0.282 58	0.577 56	0.380 53	0.182 60	0.107 64	0.430 59	0.001 65	0.422 43	0.057 60	0.179 58	0.162 61	0.070 61	0.229 56	0.511 52	0.161 55	0.491 53	0.313 55	0.650 64	0.162 57
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
SegGroup_ins	0.246 63	0.556 57	0.335 59	0.062 70	0.115 63	0.490 55	0.000 68	0.297 60	0.018 67	0.186 56	0.142 62	0.083 59	0.233 55	0.216 66	0.153 59	0.469 55	0.251 61	0.744 55	0.083 66
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
UNet-backbone	0.161 65	0.519 58	0.259 65	0.084 64	0.059 67	0.325 69	0.002 63	0.093 70	0.009 69	0.077 68	0.064 67	0.045 63	0.044 71	0.161 68	0.045 67	0.331 66	0.180 65	0.566 65	0.033 75

3D-BoNet	0.253 61	0.519 58	0.324 61	0.251 54	0.137 61	0.345 67	0.031 46	0.419 44	0.069 56	0.162 61	0.131 63	0.052 62	0.202 60	0.338 62	0.147 60	0.301 68	0.303 59	0.651 63	0.178 56
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
RWSeg	0.348 53	0.475 60	0.456 51	0.320 49	0.275 43	0.476 56	0.020 53	0.491 35	0.056 61	0.212 51	0.320 45	0.261 38	0.302 52	0.520 49	0.182 52	0.557 44	0.285 60	0.867 36	0.197 52

One_Thing_One_Click	0.326 55	0.472 61	0.361 56	0.232 56	0.183 56	0.555 52	0.000 68	0.498 34	0.038 63	0.195 55	0.226 57	0.362 26	0.168 62	0.469 56	0.251 42	0.553 45	0.335 52	0.846 44	0.117 63
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
MASC	0.254 60	0.463 62	0.249 66	0.113 62	0.167 57	0.412 62	0.000 67	0.374 50	0.073 55	0.173 60	0.243 54	0.130 55	0.228 57	0.368 60	0.160 56	0.356 63	0.208 63	0.711 60	0.136 61
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
3D-MPA	0.355 51	0.457 63	0.484 47	0.299 50	0.277 42	0.591 50	0.047 43	0.332 52	0.212 36	0.217 48	0.278 49	0.193 49	0.413 43	0.410 58	0.195 49	0.574 40	0.352 47	0.849 41	0.213 51
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
3D-SIS	0.161 65	0.407 64	0.155 72	0.068 67	0.043 71	0.346 66	0.001 64	0.134 67	0.005 70	0.088 65	0.106 66	0.037 65	0.135 66	0.321 63	0.028 71	0.339 64	0.116 72	0.466 68	0.093 65
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
SPG_WSIS	0.251 62	0.380 65	0.274 64	0.289 51	0.144 59	0.413 61	0.000 68	0.311 56	0.065 57	0.113 64	0.130 64	0.029 67	0.204 59	0.388 59	0.108 64	0.459 57	0.311 56	0.769 53	0.127 62

R-PointNet	0.158 67	0.356 66	0.173 70	0.113 63	0.140 60	0.359 63	0.012 56	0.023 73	0.039 62	0.134 63	0.123 65	0.008 71	0.089 67	0.149 69	0.117 63	0.221 70	0.128 70	0.563 66	0.094 64

tmp	0.113 72	0.333 67	0.151 73	0.056 71	0.053 69	0.344 68	0.000 68	0.105 69	0.016 68	0.049 72	0.035 71	0.020 69	0.053 68	0.048 72	0.013 73	0.183 74	0.173 66	0.344 72	0.054 72

SemRegionNet-20cls	0.121 69	0.296 68	0.203 68	0.071 66	0.058 68	0.349 65	0.000 68	0.150 66	0.019 66	0.054 71	0.034 72	0.017 70	0.052 69	0.042 73	0.013 74	0.209 72	0.183 64	0.371 69	0.057 71

Sem_Recon_ins	0.098 73	0.295 69	0.187 69	0.015 75	0.036 72	0.213 74	0.005 61	0.038 72	0.003 71	0.056 70	0.037 70	0.036 66	0.015 74	0.051 71	0.044 68	0.209 73	0.098 74	0.354 71	0.071 68

3D-BEVIS	0.117 70	0.250 70	0.308 63	0.020 74	0.009 76	0.269 72	0.006 60	0.008 74	0.029 64	0.037 74	0.014 75	0.003 73	0.036 72	0.147 70	0.042 69	0.381 62	0.118 71	0.362 70	0.069 69
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.
PanopticFusion-inst	0.214 64	0.250 70	0.330 60	0.275 53	0.103 65	0.228 73	0.000 68	0.345 51	0.024 65	0.088 66	0.203 60	0.186 50	0.167 63	0.367 61	0.125 62	0.221 71	0.112 73	0.666 62	0.162 58
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
Hier3D	0.117 70	0.222 72	0.161 71	0.054 72	0.027 73	0.289 70	0.000 68	0.124 68	0.001 74	0.079 67	0.061 68	0.027 68	0.141 65	0.240 65	0.005 75	0.310 67	0.129 69	0.153 75	0.081 67
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
MaskRCNN 2d->3d Proj	0.022 76	0.185 73	0.000 76	0.000 76	0.015 74	0.000 76	0.000 66	0.006 76	0.000 75	0.010 76	0.006 76	0.107 57	0.012 75	0.000 75	0.002 76	0.027 76	0.004 76	0.022 76	0.001 76

Region-18class	0.146 68	0.175 74	0.321 62	0.080 65	0.062 66	0.357 64	0.000 68	0.307 57	0.002 72	0.066 69	0.044 69	0.000 75	0.018 73	0.036 74	0.054 66	0.447 60	0.133 68	0.472 67	0.060 70

ASIS	0.085 74	0.037 75	0.080 75	0.066 69	0.047 70	0.282 71	0.000 68	0.052 71	0.002 73	0.047 73	0.026 73	0.001 74	0.046 70	0.194 67	0.031 70	0.264 69	0.140 67	0.167 74	0.047 74

Sgpn_scannet	0.049 75	0.023 76	0.134 74	0.031 73	0.013 75	0.144 75	0.006 58	0.008 75	0.000 75	0.028 75	0.017 74	0.003 72	0.009 76	0.000 75	0.021 72	0.122 75	0.095 75	0.175 73	0.054 73

3D Semantic Instance Benchmark