Presenting the ScanNet200 Benchmark

We present the ScanNet200 benchmark, which studies an order of magnitude more class categories than previous version of ScanNet. The scene geometry is shared within the two tasks, but the parsing of surface annotation allows for a larger vocabulary and more realistic setting for in the wild 3D understanding methods.

The ScanNet200 benchmark includes both finer-grained categories as well as a large number of previously unaddressed classes. This induces a much more challenging setting regarding the diversity of naturally observed semantic classes seen in the raw ScanNet RGB-D observations, where the data also reflects naturally encountered class imbalances. The difference in category frequencies between ScanNet and ScanNet200 can be seen in the Figure above.

ScanNet200 Benchmark

This table lists the benchmark results for the ScanNet200 3D semantic label scenario.




Method Infoavg iouhead ioucommon ioutail ioualarm clockarmchairbackpackbagballbarbasketbathroom cabinetbathroom counterbathroom stallbathroom stall doorbathroom vanitybathtubbedbenchbicyclebinblackboardblanketblindsboardbookbookshelfbottlebowlboxbroombucketbulletin boardcabinetcalendarcandlecartcase of water bottlescd caseceilingceiling lightchairclockclosetcloset doorcloset rodcloset wallclothesclothes dryercoat rackcoffee kettlecoffee makercoffee tablecolumncomputer towercontainercopiercouchcountercratecupcurtaincushiondecorationdeskdining tabledish rackdishwasherdividerdoordoorframedresserdumbbelldustpanend tablefanfile cabinetfire alarmfire extinguisherfireplacefloorfolded chairfurnitureguitarguitar casehair dryerhandicap barhatheadphonesironing boardjacketkeyboardkeyboard pianokitchen cabinetkitchen counterladderlamplaptoplaundry basketlaundry detergentlaundry hamperledgelightlight switchluggagemachinemailboxmatmattressmicrowavemini fridgemirrormonitormousemusic standnightstandobjectoffice chairottomanovenpaperpaper bagpaper cutterpaper towel dispenserpaper towel rollpersonpianopicturepillarpillowpipeplantplateplungerposterpotted plantpower outletpower stripprinterprojectorprojector screenpurserackradiatorrailrange hoodrecycling binrefrigeratorscaleseatshelfshoeshowershower curtainshower curtain rodshower doorshower floorshower headshower wallsignsinksoap dishsoap dispensersofa chairspeakerstair railstairsstandstoolstorage binstorage containerstorage organizerstovestructurestuffed animalsuitcasetabletelephonetissue boxtoastertoaster oventoilettoilet papertoilet paper dispensertoilet paper holdertoilet seat cover dispensertoweltrash bintrash cantraytubetvtv standvacuum cleanerventwallwardrobewashing machinewater bottlewater coolerwater pitcherwhiteboardwindowwindowsill
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
CSC-Pretrainpermissive0.249 130.455 130.171 120.079 130.418 120.059 120.186 80.000 10.000 30.000 10.335 80.250 110.316 120.766 70.697 130.142 100.170 100.003 20.553 100.112 70.097 10.201 130.186 100.476 130.081 120.000 60.216 130.000 10.000 30.001 130.314 130.000 70.000 10.055 110.000 20.832 130.094 30.659 110.002 40.076 70.310 130.293 130.664 120.000 10.000 20.175 130.634 40.130 20.552 130.686 130.700 130.076 60.110 110.770 130.000 10.000 90.430 130.000 70.319 110.166 110.542 130.327 120.205 120.332 120.052 120.375 90.444 130.000 50.012 130.930 130.203 20.000 10.000 90.046 80.175 100.413 120.592 100.471 120.299 110.152 130.340 120.247 130.000 30.000 10.225 110.058 30.037 30.000 80.207 10.862 120.014 100.548 90.033 120.233 120.816 120.000 80.000 10.542 110.123 40.121 10.019 20.000 10.000 70.463 120.454 130.045 130.128 130.557 120.235 100.441 120.063 90.484 130.000 40.308 130.000 10.000 50.000 30.318 130.000 20.000 50.000 10.545 120.543 100.164 120.734 70.000 80.000 10.215 130.371 120.198 100.743 100.205 120.062 110.000 80.079 100.000 10.683 120.547 120.142 70.000 80.441 70.579 120.000 10.464 110.098 70.041 10.000 10.590 120.000 20.000 10.373 90.494 100.174 110.105 120.001 130.895 120.222 120.537 90.307 120.180 50.625 100.000 10.000 90.591 130.609 110.398 110.000 10.766 130.014 130.638 130.000 10.377 100.004 90.206 110.609 130.465 5
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
LGroundpermissive0.272 110.485 110.184 110.106 110.476 80.077 80.218 60.000 10.000 30.000 10.547 10.295 90.540 40.746 90.745 110.058 120.112 120.005 10.658 70.077 130.000 70.322 110.178 120.512 100.190 90.199 10.277 110.000 10.000 30.173 50.399 80.000 70.000 10.039 120.000 20.858 110.085 60.676 80.002 40.103 40.498 70.323 100.703 100.000 10.000 20.296 110.549 80.216 10.702 50.768 100.718 100.028 90.092 120.786 120.000 10.000 90.453 120.022 50.251 130.252 80.572 110.348 110.321 70.514 60.063 110.279 120.552 110.000 50.019 120.932 110.132 120.000 10.000 90.000 110.156 130.457 110.623 90.518 100.265 120.358 80.381 110.395 110.000 30.000 10.127 130.012 50.051 10.000 80.000 30.886 100.014 100.437 130.179 40.244 110.826 110.000 80.000 10.599 80.136 10.085 30.000 40.000 10.000 70.565 90.612 100.143 40.207 110.566 110.232 110.446 110.127 20.708 110.000 40.384 70.000 10.000 50.000 30.402 100.000 20.059 30.000 10.525 130.566 90.229 100.659 110.000 80.000 10.265 110.446 100.147 120.720 130.597 60.066 100.000 80.187 50.000 10.726 90.467 130.134 90.000 80.413 110.629 90.000 10.363 120.055 80.022 20.000 10.626 90.000 20.000 10.323 110.479 130.154 120.117 110.028 120.901 110.243 110.415 120.295 130.143 60.610 120.000 10.000 90.777 80.397 130.324 120.000 10.778 110.179 80.702 120.000 10.274 130.404 10.233 80.622 110.398 7
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
PTv3 ScanNet2000.393 20.592 20.330 10.216 20.520 20.109 30.108 120.000 10.337 10.000 10.310 100.394 70.494 100.753 80.848 10.256 30.717 40.000 30.842 20.192 30.065 30.449 70.346 20.546 50.190 90.000 60.384 50.000 10.000 30.218 20.505 10.791 20.000 10.136 20.000 20.903 20.073 100.687 50.000 60.168 10.551 40.387 60.941 20.000 10.000 20.397 90.654 30.000 90.714 40.759 110.752 60.118 40.264 30.926 20.000 10.048 40.575 30.000 70.597 10.366 10.755 10.469 10.474 20.798 10.140 70.617 10.692 50.000 50.592 20.971 20.188 30.000 10.133 60.593 10.349 10.650 30.717 60.699 20.455 20.790 10.523 30.636 10.301 10.000 10.622 20.000 80.017 110.259 30.000 30.921 30.337 10.733 20.210 20.514 20.860 70.407 10.000 10.688 20.109 70.000 100.000 40.000 10.151 30.671 50.782 10.115 90.641 10.903 10.349 10.616 30.088 50.832 40.000 40.480 20.000 10.428 10.000 30.497 80.000 20.000 50.000 10.662 30.690 20.612 10.828 10.575 10.000 10.404 50.644 10.325 40.887 30.728 10.009 120.134 60.026 130.000 10.761 20.731 10.172 40.077 20.528 40.727 40.000 10.603 40.220 30.022 20.000 10.740 10.000 20.000 10.661 30.586 10.566 20.436 40.531 20.978 20.457 10.708 30.583 40.141 70.748 20.000 10.026 20.822 20.871 30.879 50.000 10.851 20.405 20.914 10.000 10.682 20.000 110.281 20.738 20.463 6
Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, Hengshuang Zhao: Point Transformer V3: Simpler, Faster, Stronger. CVPR 2024 (Oral)
OA-CNN-L_ScanNet2000.333 70.558 30.269 70.124 90.448 110.080 70.272 30.000 10.000 30.000 10.342 60.515 20.524 60.713 120.789 60.158 90.384 80.000 30.806 40.125 50.000 70.496 50.332 50.498 120.227 60.024 30.474 10.000 10.003 20.071 70.487 20.000 70.000 10.110 60.000 20.876 60.013 130.703 20.000 60.076 70.473 90.355 80.906 50.000 10.000 20.476 50.706 10.000 90.672 90.835 90.748 70.015 120.223 50.860 70.000 10.000 90.572 50.000 70.509 60.313 60.662 30.398 100.396 40.411 110.276 20.527 20.711 40.000 50.076 90.946 50.166 50.000 10.022 70.160 50.183 90.493 90.699 70.637 40.403 40.330 90.406 90.526 50.024 20.000 10.392 90.000 80.016 120.000 80.196 20.915 50.112 80.557 70.197 30.352 80.877 30.000 80.000 10.592 100.103 90.000 100.067 10.000 10.089 40.735 40.625 80.130 80.568 50.836 60.271 40.534 70.043 110.799 70.001 30.445 40.000 10.000 50.024 20.661 30.000 20.262 20.000 10.591 60.517 110.373 70.788 60.021 70.000 10.455 20.517 70.320 50.823 80.200 130.001 130.150 50.100 80.000 10.736 70.668 60.103 110.052 40.662 20.720 50.000 10.602 50.112 50.002 40.000 10.637 70.000 20.000 10.621 80.569 30.398 70.412 50.234 80.949 50.363 30.492 110.495 70.251 40.665 70.000 10.001 80.805 40.833 50.794 80.000 10.821 40.314 50.843 90.000 10.560 70.245 30.262 40.713 30.370 10
BFANet ScanNet200permissive0.360 30.553 50.293 30.193 30.483 70.096 40.266 40.000 10.000 30.000 10.298 110.255 100.661 10.810 50.810 20.194 70.785 30.000 30.000 130.161 40.000 70.494 60.382 10.574 30.258 30.000 60.372 70.000 10.000 30.043 100.436 60.000 70.000 10.239 10.000 20.901 30.105 10.689 30.025 30.128 20.614 20.436 10.493 130.000 10.000 20.526 40.546 90.109 40.651 110.953 30.753 50.101 50.143 90.897 30.000 10.431 10.469 110.000 70.522 50.337 40.661 50.459 20.409 30.666 40.102 100.508 40.757 30.000 50.060 100.970 30.497 10.000 10.376 20.511 20.262 40.688 20.921 10.617 60.321 100.590 30.491 50.556 30.000 30.000 10.481 30.093 10.043 20.284 20.000 30.875 110.135 60.669 40.124 90.394 50.849 100.298 20.000 10.476 130.088 110.042 50.000 40.000 10.254 20.653 70.741 30.215 10.573 40.852 40.266 60.654 10.056 100.835 20.000 40.492 10.000 10.000 50.000 30.612 70.000 20.000 50.000 10.616 40.469 130.460 40.698 100.516 20.000 10.378 60.563 30.476 20.863 40.574 70.330 50.000 80.282 30.000 10.760 30.710 20.233 10.000 80.641 30.814 10.000 10.585 70.053 90.000 50.000 10.629 80.000 20.000 10.678 20.528 90.534 30.129 100.596 10.973 30.264 80.772 20.526 60.139 90.707 30.000 10.000 90.764 100.591 120.848 60.000 10.827 30.338 30.806 100.000 10.568 60.151 60.358 10.659 80.510 4
ALS-MinkowskiNetcopyleft0.414 10.610 10.322 20.271 10.542 10.153 10.159 90.000 10.000 30.000 10.404 30.503 30.532 50.672 130.804 40.285 10.888 10.000 30.900 10.226 10.087 20.598 30.342 30.671 10.217 80.087 20.449 20.000 10.000 30.253 10.477 41.000 10.000 10.118 30.000 20.905 10.071 110.710 10.076 10.047 120.665 10.376 70.981 10.000 10.000 20.466 60.632 50.113 30.769 10.956 20.795 10.031 80.314 10.936 10.000 10.390 20.601 10.000 70.458 70.366 10.719 20.440 40.564 10.699 30.314 10.464 50.784 10.200 10.283 40.973 10.142 80.000 10.250 40.285 40.220 50.718 10.752 40.723 10.460 10.248 120.475 60.463 100.000 30.000 10.446 60.021 40.025 70.285 10.000 30.972 10.149 50.769 10.230 10.535 10.879 20.252 40.000 10.693 10.129 20.000 100.000 40.000 10.447 10.958 10.662 60.159 20.598 20.780 100.344 20.646 20.106 40.893 10.135 10.455 30.000 10.194 30.259 10.726 20.475 10.000 50.000 10.741 10.865 10.571 20.817 30.445 30.000 10.506 10.630 20.230 80.916 10.728 10.635 11.000 10.252 40.000 10.804 10.697 30.137 80.043 50.717 10.807 20.000 10.510 100.245 10.000 50.000 10.709 20.000 20.000 10.703 10.572 20.646 10.223 90.531 20.984 10.397 20.813 10.798 10.135 110.800 10.000 10.097 10.832 10.752 70.842 70.000 10.852 10.149 90.846 80.000 10.666 40.359 20.252 60.777 10.690 2
PonderV2 ScanNet2000.346 40.552 60.270 60.175 50.497 60.070 100.239 50.000 10.000 30.000 10.232 130.412 60.584 20.842 30.804 40.212 60.540 60.000 30.433 120.106 80.000 70.590 40.290 80.548 40.243 50.000 60.356 90.000 10.000 30.062 80.398 90.441 60.000 10.104 80.000 20.888 40.076 90.682 60.030 20.094 50.491 80.351 90.869 80.000 10.063 10.403 80.700 20.000 90.660 100.881 50.761 20.050 70.186 60.852 90.000 10.007 70.570 60.100 20.565 20.326 50.641 80.431 50.290 100.621 50.259 30.408 70.622 80.125 20.082 80.950 40.179 40.000 10.263 30.424 30.193 70.558 50.880 20.545 90.375 50.727 20.445 80.499 70.000 30.000 10.475 50.002 60.034 50.083 60.000 30.924 20.290 30.636 50.115 100.400 40.874 40.186 60.000 10.611 60.128 30.113 20.000 40.000 10.000 70.584 80.636 70.103 100.385 70.843 50.283 30.603 50.080 60.825 60.000 40.377 80.000 10.000 50.000 30.457 90.000 20.000 50.000 10.574 100.608 70.481 30.792 40.394 40.000 10.357 80.503 90.261 70.817 90.504 100.304 60.472 40.115 70.000 10.750 50.677 50.202 20.000 80.509 50.729 30.000 10.519 90.000 120.000 50.000 10.620 100.000 20.000 10.660 50.560 50.486 40.384 60.346 60.952 40.247 100.667 40.436 80.269 30.691 50.000 10.010 40.787 60.889 20.880 40.000 10.810 60.336 40.860 60.000 10.606 50.009 70.248 70.681 50.392 8
Haoyi Zhu, Honghui Yang, Xiaoyang Wu, Di Huang, Sha Zhang, Xianglong He, Tong He, Hengshuang Zhao, Chunhua Shen, Yu Qiao, Wanli Ouyang: PonderV2: Pave the Way for 3D Foundataion Model with A Universal Pre-training Paradigm.
L3DETR-ScanNet_2000.336 60.533 90.279 40.155 60.508 40.073 90.101 130.000 10.058 20.000 10.294 120.233 120.548 30.927 10.788 70.264 20.463 70.000 30.638 80.098 110.014 50.411 90.226 90.525 90.225 70.010 40.397 40.000 10.000 30.192 40.380 100.598 40.000 10.117 40.000 20.883 50.082 70.689 30.000 60.032 130.549 50.417 40.910 40.000 10.000 20.448 70.613 70.000 90.697 60.960 10.759 30.158 20.293 20.883 50.000 10.312 30.583 20.079 40.422 100.068 130.660 60.418 60.298 80.430 100.114 80.526 30.776 20.051 30.679 10.946 50.152 60.000 10.183 50.000 110.211 60.511 80.409 120.565 80.355 60.448 50.512 40.557 20.000 30.000 10.420 70.000 80.007 130.104 40.000 30.125 130.330 20.514 110.146 80.321 90.860 70.174 70.000 10.629 40.075 120.000 100.000 40.000 10.002 60.671 50.712 40.141 50.339 80.856 30.261 80.529 80.067 80.835 20.000 40.369 100.000 10.259 20.000 30.629 40.000 20.487 10.000 10.579 90.646 30.107 130.720 90.122 60.000 10.333 100.505 80.303 60.908 20.503 110.565 20.074 70.324 10.000 10.740 60.661 70.109 100.000 80.427 90.563 130.000 10.579 80.108 60.000 50.000 10.664 40.000 20.000 10.641 60.539 70.416 50.515 20.256 70.940 90.312 40.209 130.620 20.138 100.636 90.000 10.000 90.775 90.861 40.765 90.000 10.801 80.119 110.860 60.000 10.687 10.001 100.192 120.679 70.699 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12
PPT-SpUNet-F.T.0.332 80.556 40.270 50.123 100.519 30.091 50.349 20.000 10.000 30.000 10.339 70.383 80.498 90.833 40.807 30.241 40.584 50.000 30.755 50.124 60.000 70.608 20.330 60.530 80.314 10.000 60.374 60.000 10.000 30.197 30.459 50.000 70.000 10.117 40.000 20.876 60.095 20.682 60.000 60.086 60.518 60.433 20.930 30.000 10.000 20.563 30.542 100.077 60.715 30.858 70.756 40.008 130.171 80.874 60.000 10.039 50.550 70.000 70.545 40.256 70.657 70.453 30.351 60.449 90.213 40.392 80.611 90.000 50.037 110.946 50.138 100.000 10.000 90.063 70.308 20.537 60.796 30.673 30.323 90.392 70.400 100.509 60.000 30.000 10.649 10.000 80.023 80.000 80.000 30.914 60.002 120.506 120.163 70.359 70.872 50.000 80.000 10.623 50.112 50.001 90.000 40.000 10.021 50.753 20.565 120.150 30.579 30.806 80.267 50.616 30.042 120.783 90.000 40.374 90.000 10.000 50.000 30.620 60.000 20.000 50.000 10.572 110.634 40.350 80.792 40.000 80.000 10.376 70.535 50.378 30.855 50.672 30.074 90.000 80.185 60.000 10.727 80.660 80.076 130.000 80.432 80.646 70.000 10.594 60.006 110.000 50.000 10.658 50.000 20.000 10.661 30.549 60.300 100.291 80.045 100.942 80.304 50.600 70.572 50.135 110.695 40.000 10.008 60.793 50.942 10.899 20.000 10.816 50.181 70.897 20.000 10.679 30.223 40.264 30.691 40.345 11
Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao: Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training. CVPR 2024
OctFormer ScanNet200permissive0.326 90.539 80.265 80.131 80.499 50.110 20.522 10.000 10.000 30.000 10.318 90.427 50.455 110.743 100.765 90.175 80.842 20.000 30.828 30.204 20.033 40.429 80.335 40.601 20.312 20.000 60.357 80.000 10.000 30.047 90.423 70.000 70.000 10.105 70.000 20.873 80.079 80.670 90.000 60.117 30.471 100.432 30.829 90.000 10.000 20.584 20.417 130.089 50.684 80.837 80.705 120.021 110.178 70.892 40.000 10.028 60.505 90.000 70.457 80.200 100.662 30.412 80.244 110.496 70.000 130.451 60.626 70.000 50.102 70.943 80.138 100.000 10.000 90.149 60.291 30.534 70.722 50.632 50.331 80.253 110.453 70.487 80.000 30.000 10.479 40.000 80.022 90.000 80.000 30.900 70.128 70.684 30.164 60.413 30.854 90.000 80.000 10.512 120.074 130.003 80.000 40.000 10.000 70.469 110.613 90.132 70.529 60.871 20.227 120.582 60.026 130.787 80.000 40.339 110.000 10.000 50.000 30.626 50.000 20.029 40.000 10.587 70.612 60.411 60.724 80.000 80.000 10.407 40.552 40.513 10.849 60.655 40.408 30.000 80.296 20.000 10.686 110.645 100.145 60.022 60.414 100.633 80.000 10.637 10.224 20.000 50.000 10.650 60.000 20.000 10.622 70.535 80.343 80.483 30.230 90.943 70.289 60.618 60.596 30.140 80.679 60.000 10.022 30.783 70.620 100.906 10.000 10.806 70.137 100.865 30.000 10.378 90.000 110.168 130.680 60.227 12
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
CeCo0.340 50.551 70.247 90.181 40.475 90.057 130.142 100.000 10.000 30.000 10.387 40.463 40.499 80.924 20.774 80.213 50.257 90.000 30.546 110.100 90.006 60.615 10.177 130.534 60.246 40.000 60.400 30.000 10.338 10.006 120.484 30.609 30.000 10.083 90.000 20.873 80.089 50.661 100.000 60.048 110.560 30.408 50.892 60.000 10.000 20.586 10.616 60.000 90.692 70.900 40.721 80.162 10.228 40.860 70.000 10.000 90.575 30.083 30.550 30.347 30.624 90.410 90.360 50.740 20.109 90.321 110.660 60.000 50.121 50.939 90.143 70.000 10.400 10.003 90.190 80.564 40.652 80.615 70.421 30.304 100.579 10.547 40.000 30.000 10.296 100.000 80.030 60.096 50.000 30.916 40.037 90.551 80.171 50.376 60.865 60.286 30.000 10.633 30.102 100.027 60.011 30.000 10.000 70.474 100.742 20.133 60.311 90.824 70.242 90.503 100.068 70.828 50.000 40.429 50.000 10.063 40.000 30.781 10.000 20.000 50.000 10.665 20.633 50.450 50.818 20.000 80.000 10.429 30.532 60.226 90.825 70.510 90.377 40.709 20.079 100.000 10.753 40.683 40.102 120.063 30.401 120.620 100.000 10.619 20.000 120.000 50.000 10.595 110.000 20.000 10.345 100.564 40.411 60.603 10.384 50.945 60.266 70.643 50.367 100.304 10.663 80.000 10.010 40.726 110.767 60.898 30.000 10.784 90.435 10.861 50.000 10.447 80.000 110.257 50.656 90.377 9
Zhisheng Zhong, Jiequan Cui, Yibo Yang, Xiaoyang Wu, Xiaojuan Qi, Xiangyu Zhang, Jiaya Jia: Understanding Imbalanced Semantic Segmentation Through Neural Collapse. CVPR 2023
AWCS0.305 100.508 100.225 100.142 70.463 100.063 110.195 70.000 10.000 30.000 10.467 20.551 10.504 70.773 60.764 100.142 100.029 130.000 30.626 90.100 90.000 70.360 100.179 110.507 110.137 110.006 50.300 100.000 10.000 30.172 60.364 110.512 50.000 10.056 100.000 20.865 100.093 40.634 130.000 60.071 90.396 110.296 120.876 70.000 10.000 20.373 100.436 120.063 80.749 20.877 60.721 80.131 30.124 100.804 110.000 10.000 90.515 80.010 60.452 90.252 80.578 100.417 70.179 130.484 80.171 50.337 100.606 100.000 50.115 60.937 100.142 80.000 10.008 80.000 110.157 120.484 100.402 130.501 110.339 70.553 40.529 20.478 90.000 30.000 10.404 80.001 70.022 90.077 70.000 30.894 90.219 40.628 60.093 110.305 100.886 10.233 50.000 10.603 70.112 50.023 70.000 40.000 10.000 70.741 30.664 50.097 110.253 100.782 90.264 70.523 90.154 10.707 120.000 40.411 60.000 10.000 50.000 30.332 120.000 20.000 50.000 10.602 50.595 80.185 110.656 120.159 50.000 10.355 90.424 110.154 110.729 110.516 80.220 80.620 30.084 90.000 10.707 100.651 90.173 30.014 70.381 130.582 110.000 10.619 20.049 100.000 50.000 10.702 30.000 20.000 10.302 120.489 110.317 90.334 70.392 40.922 100.254 90.533 100.394 90.129 130.613 110.000 10.000 90.820 30.649 90.749 100.000 10.782 100.282 60.863 40.000 10.288 120.006 80.220 90.633 100.542 3
Minkowski 34Dpermissive0.253 120.463 120.154 130.102 120.381 130.084 60.134 110.000 10.000 30.000 10.386 50.141 130.279 130.737 110.703 120.014 130.164 110.000 30.663 60.092 120.000 70.224 120.291 70.531 70.056 130.000 60.242 120.000 10.000 30.013 110.331 120.000 70.000 10.035 130.001 10.858 110.059 120.650 120.000 60.056 100.353 120.299 110.670 110.000 10.000 20.284 120.484 110.071 70.594 120.720 120.710 110.027 100.068 130.813 100.000 10.005 80.492 100.164 10.274 120.111 120.571 120.307 130.293 90.307 130.150 60.163 130.531 120.002 40.545 30.932 110.093 130.000 10.000 90.002 100.159 110.368 130.581 110.440 130.228 130.406 60.282 130.294 120.000 30.000 10.189 120.060 20.036 40.000 80.000 30.897 80.000 130.525 100.025 130.205 130.771 130.000 80.000 10.593 90.108 80.044 40.000 40.000 10.000 70.282 130.589 110.094 120.169 120.466 130.227 120.419 130.125 30.757 100.002 20.334 120.000 10.000 50.000 30.357 110.000 20.000 50.000 10.582 80.513 120.337 90.612 130.000 80.000 10.250 120.352 130.136 130.724 120.655 40.280 70.000 80.046 120.000 10.606 130.559 110.159 50.102 10.445 60.655 60.000 10.310 130.117 40.000 50.000 10.581 130.026 10.000 10.265 130.483 120.084 130.097 130.044 110.865 130.142 130.588 80.351 110.272 20.596 130.000 10.003 70.622 120.720 80.096 130.000 10.771 120.016 120.772 110.000 10.302 110.194 50.214 100.621 120.197 13
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019


This table lists the benchmark results for the ScanNet200 3D semantic instance scenario.




Method Infoavg aphead apcommon aptail apalarm clockarmchairbackpackbagballbarbasketbathroom cabinetbathroom counterbathroom stallbathroom stall doorbathroom vanitybathtubbedbenchbicyclebinblackboardblanketblindsboardbookbookshelfbottlebowlboxbroombucketbulletin boardcabinetcalendarcandlecartcase of water bottlescd caseceilingceiling lightchairclockclosetcloset doorcloset rodcloset wallclothesclothes dryercoat rackcoffee kettlecoffee makercoffee tablecolumncomputer towercontainercopiercouchcountercratecupcurtaincushiondecorationdeskdining tabledish rackdishwasherdividerdoordoorframedresserdumbbelldustpanend tablefanfile cabinetfire alarmfire extinguisherfireplacefolded chairfurnitureguitarguitar casehair dryerhandicap barhatheadphonesironing boardjacketkeyboardkeyboard pianokitchen cabinetkitchen counterladderlamplaptoplaundry basketlaundry detergentlaundry hamperledgelightlight switchluggagemachinemailboxmatmattressmicrowavemini fridgemirrormonitormousemusic standnightstandobjectoffice chairottomanovenpaperpaper bagpaper cutterpaper towel dispenserpaper towel rollpersonpianopicturepillarpillowpipeplantplateplungerposterpotted plantpower outletpower stripprinterprojectorprojector screenpurserackradiatorrailrange hoodrecycling binrefrigeratorscaleseatshelfshoeshowershower curtainshower curtain rodshower doorshower floorshower headshower wallsignsinksoap dishsoap dispensersofa chairspeakerstair railstairsstandstoolstorage binstorage containerstorage organizerstovestructurestuffed animalsuitcasetabletelephonetissue boxtoastertoaster oventoilettoilet papertoilet paper dispensertoilet paper holdertoilet seat cover dispensertoweltrash bintrash cantraytubetvtv standvacuum cleanerventwardrobewashing machinewater bottlewater coolerwater pitcherwhiteboardwindowwindowsill
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
TD3D Scannet200permissive0.211 20.332 20.177 20.103 20.337 20.036 20.222 40.000 10.000 10.000 10.031 10.342 10.093 40.852 10.452 40.559 20.000 20.004 20.000 30.039 10.000 20.309 20.047 40.380 20.028 20.000 10.080 20.000 10.000 20.147 10.192 30.000 20.000 10.083 10.000 10.395 20.039 40.662 10.000 10.000 20.074 10.135 10.296 20.000 20.000 10.231 50.646 10.139 30.633 31.000 10.705 10.048 10.088 20.439 20.184 20.039 20.266 20.551 20.260 30.026 50.463 20.046 30.252 20.249 30.083 20.372 10.411 10.000 20.414 10.323 10.000 10.052 20.000 10.157 10.278 20.278 20.237 20.015 20.321 20.253 10.060 40.000 10.000 10.272 20.008 10.169 20.032 20.000 10.404 10.356 20.283 20.073 30.028 50.617 20.038 20.000 10.494 20.037 20.215 10.083 20.000 20.003 20.486 30.694 10.000 20.040 40.083 40.219 50.209 20.007 10.483 10.000 20.125 40.000 10.150 20.014 10.544 10.000 10.000 20.000 10.260 50.143 50.200 10.610 30.028 20.032 10.145 10.059 20.046 40.740 20.806 10.543 20.000 20.108 20.008 10.222 50.669 20.456 10.074 10.224 10.586 10.006 20.451 20.000 10.002 10.889 10.282 20.000 10.000 10.252 20.413 20.111 20.074 20.240 10.893 10.266 20.144 30.293 20.281 20.604 20.000 10.000 20.379 50.963 10.250 40.000 10.160 10.420 20.000 10.343 30.207 20.079 50.315 20.052 2
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
Mask3D Scannet2000.278 10.383 10.263 10.168 10.506 10.068 10.083 50.000 10.000 10.000 10.023 20.149 40.302 10.778 30.647 10.569 10.500 10.031 10.014 20.027 20.173 10.311 10.195 10.351 30.258 10.000 10.082 10.000 10.003 10.037 20.391 11.000 10.000 10.014 20.000 10.572 10.573 10.661 20.000 10.003 10.005 40.082 40.349 10.028 10.000 10.605 10.515 30.509 10.711 11.000 10.665 30.015 20.107 10.402 40.201 10.083 10.304 10.759 10.491 10.378 10.572 10.119 10.277 10.013 50.089 10.283 20.411 20.267 10.006 30.156 20.000 10.116 10.000 10.105 30.556 10.514 10.396 10.275 10.323 10.215 20.380 10.000 10.000 10.356 10.005 20.208 10.325 10.000 10.050 40.400 10.561 10.258 10.179 10.722 10.147 10.000 10.586 10.063 10.015 20.139 10.016 10.028 10.708 10.418 20.016 10.048 30.500 10.489 10.349 10.001 20.475 20.086 10.365 10.000 10.500 10.000 20.323 30.000 10.222 10.000 10.497 10.626 10.044 30.795 10.556 10.008 20.121 40.265 10.667 10.789 10.568 20.579 10.444 10.176 10.004 20.474 10.752 10.233 20.014 20.002 40.570 20.007 10.377 50.000 10.000 20.000 20.337 10.000 10.000 10.384 10.465 10.287 10.085 10.048 20.816 50.467 10.810 10.377 10.415 10.744 10.000 10.004 10.724 10.778 20.590 10.000 10.032 20.441 10.000 10.377 20.391 10.427 10.321 10.192 1
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
Minkowski 34D Inst.permissive0.130 40.246 40.083 40.043 50.299 40.000 50.278 10.000 10.000 10.000 10.022 30.175 30.122 20.537 40.521 20.400 30.000 20.000 30.000 30.008 30.000 20.048 40.076 30.182 50.000 40.000 10.022 40.000 10.000 20.000 30.141 50.000 20.000 10.000 30.000 10.210 40.063 20.547 50.000 10.000 20.000 50.100 20.026 50.000 20.000 10.241 40.488 40.000 40.564 51.000 10.672 20.000 30.021 40.486 10.000 30.000 30.067 40.000 30.194 50.033 40.415 40.026 40.025 50.271 10.004 40.094 50.142 50.000 20.000 40.111 30.000 10.000 30.000 10.088 40.083 50.278 20.110 40.000 40.082 50.199 50.137 30.000 10.000 10.000 30.000 30.041 40.000 30.000 10.308 20.067 30.280 30.016 40.101 30.373 50.000 30.000 10.319 40.007 40.000 30.000 30.000 20.000 30.028 50.355 50.000 20.101 10.444 20.289 20.114 50.000 30.394 30.000 20.032 50.000 10.000 30.000 20.201 50.000 10.000 20.000 10.384 20.248 40.000 50.529 40.000 30.000 30.133 30.020 50.089 30.720 30.500 40.099 40.000 20.000 50.000 30.238 40.334 50.190 30.000 30.000 50.317 50.000 30.472 10.000 10.000 20.000 20.094 50.000 10.000 10.082 50.236 40.004 50.019 40.000 30.883 20.061 50.262 20.217 40.000 40.557 50.000 10.000 20.460 40.761 40.156 50.000 10.000 30.259 40.000 10.394 10.019 40.084 40.232 40.000 5
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
CSC-Pretrain Inst.permissive0.123 50.223 50.082 50.046 40.308 30.004 30.278 10.000 10.000 10.000 10.000 50.032 50.105 30.537 40.348 50.378 40.000 20.000 30.000 30.000 50.000 20.000 50.037 50.323 40.000 40.000 10.013 50.000 10.000 20.000 30.235 20.000 20.000 10.000 30.000 10.231 30.045 30.564 40.000 10.000 20.006 30.078 50.065 30.000 20.000 10.259 30.516 20.000 40.600 41.000 10.578 50.000 30.000 50.184 50.000 30.000 30.034 50.000 30.211 40.089 30.394 50.018 50.064 40.171 40.001 50.144 30.172 40.000 20.000 40.044 40.000 10.000 30.000 10.064 50.126 40.278 20.093 50.000 40.094 40.214 30.011 50.000 10.000 10.000 30.000 30.022 50.000 30.000 10.275 30.000 40.275 40.000 50.098 40.407 40.000 30.000 10.250 50.007 50.000 30.000 30.000 20.000 30.333 40.376 40.000 20.000 50.042 50.285 30.119 40.000 30.224 50.000 20.184 30.000 10.000 30.000 20.244 40.000 10.000 20.000 10.377 30.378 20.051 20.424 50.000 30.000 30.116 50.030 40.125 20.441 40.444 50.063 50.000 20.042 30.000 30.297 20.483 30.096 50.000 30.028 20.338 40.000 30.444 30.000 10.000 20.000 20.189 40.000 10.000 10.141 40.152 50.017 40.000 50.000 30.838 40.193 30.111 50.105 50.198 30.588 30.000 10.000 20.542 30.343 50.267 30.000 10.000 30.108 50.000 10.333 40.000 50.228 20.202 50.022 4
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
LGround Inst.permissive0.154 30.275 30.108 30.060 30.295 50.002 40.278 10.000 10.000 10.000 10.006 40.272 20.064 50.815 20.503 30.333 50.000 20.000 30.556 10.001 40.000 20.148 30.078 20.448 10.007 30.000 10.024 30.000 10.000 20.000 30.190 40.000 20.000 10.000 30.000 10.209 50.031 50.573 30.000 10.000 20.041 20.099 30.037 40.000 20.000 10.327 20.364 50.181 20.642 21.000 10.654 40.000 30.023 30.429 30.000 30.000 30.097 30.000 30.278 20.267 20.434 30.048 20.092 30.257 20.030 30.097 40.189 30.000 20.089 20.000 50.000 10.000 30.000 10.115 20.166 30.222 50.222 30.003 30.127 30.213 40.169 20.000 10.000 10.000 30.000 30.044 30.000 30.000 10.000 50.000 40.268 50.222 20.130 20.494 30.000 30.000 10.363 30.015 30.000 30.000 30.000 20.000 30.611 20.400 30.000 20.056 20.278 30.242 40.180 30.000 30.383 40.000 20.209 20.000 10.000 30.000 20.364 20.000 10.000 20.000 10.323 40.302 30.019 40.654 20.000 30.000 30.141 20.045 30.000 50.427 50.514 30.143 30.000 20.028 40.000 30.252 30.402 40.156 40.000 30.028 20.470 30.000 30.444 30.000 10.000 20.000 20.205 30.000 10.000 10.203 30.381 30.026 30.037 30.000 30.881 30.099 40.135 40.239 30.000 40.585 40.000 10.000 20.616 20.778 20.322 20.000 10.000 30.407 30.000 10.333 40.148 30.177 30.242 30.028 3
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild.


ScanNet Benchmark

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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
PTv3-PPT-ALCcopyleft0.798 10.911 100.812 200.854 60.770 110.856 130.555 130.943 10.660 230.735 20.979 10.606 60.492 10.792 30.934 30.841 20.819 40.716 70.947 90.906 10.822 1
PTv3 ScanNet0.794 20.941 30.813 190.851 80.782 60.890 20.597 10.916 30.696 80.713 40.979 10.635 10.384 30.793 20.907 90.821 50.790 320.696 120.967 30.903 20.805 2
Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, Hengshuang Zhao: Point Transformer V3: Simpler, Faster, Stronger. CVPR 2024 (Oral)
PonderV20.785 30.978 10.800 280.833 240.788 40.853 170.545 170.910 60.713 10.705 50.979 10.596 80.390 20.769 130.832 420.821 50.792 310.730 10.975 10.897 50.785 5
Haoyi Zhu, Honghui Yang, Xiaoyang Wu, Di Huang, Sha Zhang, Xianglong He, Tong He, Hengshuang Zhao, Chunhua Shen, Yu Qiao, Wanli Ouyang: PonderV2: Pave the Way for 3D Foundataion Model with A Universal Pre-training Paradigm.
Mix3Dpermissive0.781 40.964 20.855 10.843 170.781 70.858 120.575 60.831 340.685 140.714 30.979 10.594 90.310 270.801 10.892 170.841 20.819 40.723 40.940 140.887 70.725 25
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
Swin3Dpermissive0.779 50.861 210.818 140.836 210.790 30.875 40.576 50.905 70.704 50.739 10.969 110.611 20.349 110.756 230.958 10.702 460.805 150.708 80.916 330.898 40.801 3
TTT-KD0.773 60.646 920.818 140.809 360.774 90.878 30.581 20.943 10.687 120.704 60.978 50.607 50.336 160.775 90.912 70.838 40.823 20.694 130.967 30.899 30.794 4
Lisa Weijler, Muhammad Jehanzeb Mirza, Leon Sick, Can Ekkazan, Pedro Hermosilla: TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Models.
ResLFE_HDS0.772 70.939 40.824 60.854 60.771 100.840 310.564 100.900 90.686 130.677 130.961 170.537 320.348 120.769 130.903 110.785 110.815 70.676 230.939 150.880 120.772 9
PPT-SpUNet-Joint0.766 80.932 50.794 340.829 260.751 230.854 150.540 210.903 80.630 350.672 160.963 150.565 220.357 90.788 40.900 130.737 260.802 160.685 180.950 70.887 70.780 6
Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao: Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training. CVPR 2024
OctFormerpermissive0.766 80.925 70.808 240.849 100.786 50.846 270.566 90.876 160.690 100.674 150.960 180.576 180.226 680.753 250.904 100.777 130.815 70.722 50.923 280.877 140.776 8
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
OccuSeg+Semantic0.764 100.758 590.796 320.839 190.746 260.907 10.562 110.850 260.680 160.672 160.978 50.610 30.335 180.777 70.819 460.847 10.830 10.691 150.972 20.885 90.727 23
CU-Hybrid Net0.764 100.924 80.819 120.840 180.757 180.853 170.580 30.848 270.709 30.643 250.958 220.587 130.295 340.753 250.884 210.758 200.815 70.725 30.927 250.867 230.743 16
O-CNNpermissive0.762 120.924 80.823 70.844 160.770 110.852 190.577 40.847 290.711 20.640 290.958 220.592 100.217 740.762 180.888 180.758 200.813 110.726 20.932 230.868 220.744 15
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
DTC0.757 130.843 270.820 100.847 130.791 20.862 100.511 340.870 180.707 40.652 210.954 360.604 70.279 450.760 190.942 20.734 270.766 450.701 110.884 550.874 200.736 17
OA-CNN-L_ScanNet200.756 140.783 450.826 50.858 40.776 80.837 340.548 160.896 120.649 270.675 140.962 160.586 140.335 180.771 120.802 500.770 160.787 340.691 150.936 180.880 120.761 11
ConDaFormer0.755 150.927 60.822 80.836 210.801 10.849 220.516 310.864 230.651 260.680 120.958 220.584 160.282 420.759 210.855 320.728 290.802 160.678 200.880 600.873 210.756 13
Lunhao Duan, Shanshan Zhao, Nan Xue, Mingming Gong, Guisong Xia, Dacheng Tao: ConDaFormer : Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding. Neurips, 2023
PNE0.755 150.786 430.835 40.834 230.758 160.849 220.570 80.836 330.648 280.668 180.978 50.581 170.367 70.683 360.856 300.804 70.801 200.678 200.961 50.889 60.716 30
P. Hermosilla: Point Neighborhood Embeddings.
DMF-Net0.752 170.906 130.793 360.802 420.689 410.825 470.556 120.867 190.681 150.602 450.960 180.555 280.365 80.779 60.859 270.747 230.795 280.717 60.917 320.856 310.764 10
C.Yang, Y.Yan, W.Zhao, J.Ye, X.Yang, A.Hussain, B.Dong, K.Huang: Towards Deeper and Better Multi-view Feature Fusion for 3D Semantic Segmentation. ICONIP 2023
PointTransformerV20.752 170.742 670.809 230.872 10.758 160.860 110.552 140.891 140.610 420.687 70.960 180.559 260.304 300.766 160.926 50.767 170.797 240.644 340.942 120.876 170.722 27
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
BPNetcopyleft0.749 190.909 110.818 140.811 340.752 210.839 330.485 480.842 300.673 180.644 240.957 260.528 380.305 290.773 100.859 270.788 90.818 60.693 140.916 330.856 310.723 26
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
PointConvFormer0.749 190.793 410.790 370.807 380.750 250.856 130.524 270.881 150.588 540.642 280.977 90.591 110.274 480.781 50.929 40.804 70.796 250.642 350.947 90.885 90.715 31
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
MSP0.748 210.623 950.804 260.859 30.745 270.824 490.501 380.912 50.690 100.685 90.956 270.567 210.320 240.768 150.918 60.720 340.802 160.676 230.921 300.881 110.779 7
StratifiedFormerpermissive0.747 220.901 140.803 270.845 150.757 180.846 270.512 330.825 370.696 80.645 230.956 270.576 180.262 590.744 300.861 260.742 240.770 430.705 90.899 450.860 280.734 18
Xin Lai*, Jianhui Liu*, Li Jiang, Liwei Wang, Hengshuang Zhao, Shu Liu, Xiaojuan Qi, Jiaya Jia: Stratified Transformer for 3D Point Cloud Segmentation. CVPR 2022
Virtual MVFusion0.746 230.771 530.819 120.848 120.702 380.865 90.397 860.899 100.699 60.664 190.948 560.588 120.330 200.746 290.851 360.764 180.796 250.704 100.935 190.866 240.728 21
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
VMNetpermissive0.746 230.870 190.838 20.858 40.729 320.850 210.501 380.874 170.587 550.658 200.956 270.564 230.299 320.765 170.900 130.716 370.812 120.631 400.939 150.858 290.709 32
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)
DiffSeg3D20.745 250.725 760.814 180.837 200.751 230.831 410.514 320.896 120.674 170.684 100.960 180.564 230.303 310.773 100.820 450.713 400.798 230.690 170.923 280.875 180.757 12
Retro-FPN0.744 260.842 280.800 280.767 560.740 280.836 360.541 190.914 40.672 190.626 330.958 220.552 290.272 500.777 70.886 200.696 470.801 200.674 260.941 130.858 290.717 28
Peng Xiang*, Xin Wen*, Yu-Shen Liu, Hui Zhang, Yi Fang, Zhizhong Han: Retrospective Feature Pyramid Network for Point Cloud Semantic Segmentation. ICCV 2023
EQ-Net0.743 270.620 960.799 310.849 100.730 310.822 510.493 450.897 110.664 200.681 110.955 300.562 250.378 40.760 190.903 110.738 250.801 200.673 270.907 370.877 140.745 14
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
SAT0.742 280.860 220.765 500.819 290.769 130.848 240.533 230.829 350.663 210.631 320.955 300.586 140.274 480.753 250.896 150.729 280.760 510.666 290.921 300.855 330.733 19
LRPNet0.742 280.816 360.806 250.807 380.752 210.828 450.575 60.839 320.699 60.637 300.954 360.520 410.320 240.755 240.834 400.760 190.772 400.676 230.915 350.862 260.717 28
LargeKernel3D0.739 300.909 110.820 100.806 400.740 280.852 190.545 170.826 360.594 530.643 250.955 300.541 310.263 580.723 340.858 290.775 150.767 440.678 200.933 210.848 380.694 37
Yukang Chen*, Jianhui Liu*, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs. CVPR 2023
RPN0.736 310.776 490.790 370.851 80.754 200.854 150.491 470.866 210.596 520.686 80.955 300.536 330.342 140.624 510.869 230.787 100.802 160.628 410.927 250.875 180.704 34
MinkowskiNetpermissive0.736 310.859 230.818 140.832 250.709 360.840 310.521 290.853 250.660 230.643 250.951 460.544 300.286 400.731 320.893 160.675 560.772 400.683 190.874 670.852 360.727 23
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
IPCA0.731 330.890 150.837 30.864 20.726 330.873 50.530 260.824 380.489 880.647 220.978 50.609 40.336 160.624 510.733 590.758 200.776 380.570 660.949 80.877 140.728 21
PointTransformer++0.725 340.727 750.811 220.819 290.765 140.841 300.502 370.814 430.621 380.623 350.955 300.556 270.284 410.620 530.866 240.781 120.757 550.648 320.932 230.862 260.709 32
SparseConvNet0.725 340.647 910.821 90.846 140.721 340.869 60.533 230.754 590.603 480.614 370.955 300.572 200.325 220.710 350.870 220.724 320.823 20.628 410.934 200.865 250.683 40
MatchingNet0.724 360.812 380.812 200.810 350.735 300.834 380.495 440.860 240.572 620.602 450.954 360.512 430.280 440.757 220.845 380.725 310.780 360.606 510.937 170.851 370.700 36
INS-Conv-semantic0.717 370.751 620.759 530.812 330.704 370.868 70.537 220.842 300.609 440.608 410.953 400.534 350.293 350.616 540.864 250.719 360.793 290.640 360.933 210.845 420.663 46
PointMetaBase0.714 380.835 290.785 390.821 270.684 430.846 270.531 250.865 220.614 390.596 490.953 400.500 460.246 640.674 370.888 180.692 480.764 470.624 430.849 820.844 430.675 42
contrastBoundarypermissive0.705 390.769 560.775 440.809 360.687 420.820 540.439 740.812 440.661 220.591 510.945 640.515 420.171 920.633 480.856 300.720 340.796 250.668 280.889 520.847 390.689 38
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
ClickSeg_Semantic0.703 400.774 510.800 280.793 470.760 150.847 260.471 520.802 470.463 950.634 310.968 130.491 490.271 520.726 330.910 80.706 420.815 70.551 780.878 610.833 440.570 78
RFCR0.702 410.889 160.745 640.813 320.672 460.818 580.493 450.815 420.623 360.610 390.947 580.470 580.249 630.594 570.848 370.705 430.779 370.646 330.892 500.823 500.611 61
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
One Thing One Click0.701 420.825 330.796 320.723 630.716 350.832 400.433 760.816 400.634 330.609 400.969 110.418 840.344 130.559 690.833 410.715 380.808 140.560 720.902 420.847 390.680 41
JSENetpermissive0.699 430.881 180.762 510.821 270.667 470.800 700.522 280.792 500.613 400.607 420.935 840.492 480.205 790.576 620.853 340.691 500.758 530.652 310.872 700.828 470.649 50
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
One-Thing-One-Click0.693 440.743 660.794 340.655 860.684 430.822 510.497 430.719 690.622 370.617 360.977 90.447 710.339 150.750 280.664 750.703 450.790 320.596 560.946 110.855 330.647 51
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
PicassoNet-IIpermissive0.692 450.732 710.772 450.786 480.677 450.866 80.517 300.848 270.509 810.626 330.952 440.536 330.225 700.545 750.704 660.689 530.810 130.564 710.903 410.854 350.729 20
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
Feature_GeometricNetpermissive0.690 460.884 170.754 570.795 450.647 540.818 580.422 780.802 470.612 410.604 430.945 640.462 610.189 870.563 680.853 340.726 300.765 460.632 390.904 390.821 530.606 65
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
FusionNet0.688 470.704 810.741 680.754 600.656 490.829 430.501 380.741 640.609 440.548 590.950 500.522 400.371 50.633 480.756 540.715 380.771 420.623 440.861 780.814 560.658 47
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
Feature-Geometry Netpermissive0.685 480.866 200.748 610.819 290.645 560.794 730.450 640.802 470.587 550.604 430.945 640.464 600.201 820.554 710.840 390.723 330.732 650.602 540.907 370.822 520.603 68
VACNN++0.684 490.728 740.757 560.776 530.690 390.804 680.464 570.816 400.577 610.587 520.945 640.508 450.276 470.671 380.710 640.663 610.750 590.589 610.881 580.832 460.653 49
DGNet0.684 490.712 800.784 400.782 520.658 480.835 370.499 420.823 390.641 300.597 480.950 500.487 510.281 430.575 630.619 790.647 690.764 470.620 460.871 730.846 410.688 39
KP-FCNN0.684 490.847 260.758 550.784 500.647 540.814 610.473 510.772 530.605 460.594 500.935 840.450 690.181 900.587 580.805 490.690 510.785 350.614 470.882 570.819 540.632 57
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
Superpoint Network0.683 520.851 250.728 720.800 440.653 510.806 660.468 540.804 450.572 620.602 450.946 610.453 680.239 670.519 800.822 430.689 530.762 500.595 580.895 480.827 480.630 58
PointContrast_LA_SEM0.683 520.757 600.784 400.786 480.639 580.824 490.408 810.775 520.604 470.541 610.934 880.532 360.269 540.552 720.777 520.645 720.793 290.640 360.913 360.824 490.671 43
VI-PointConv0.676 540.770 550.754 570.783 510.621 620.814 610.552 140.758 570.571 640.557 570.954 360.529 370.268 560.530 780.682 700.675 560.719 680.603 530.888 530.833 440.665 45
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
ROSMRF3D0.673 550.789 420.748 610.763 580.635 600.814 610.407 830.747 610.581 590.573 540.950 500.484 520.271 520.607 550.754 550.649 660.774 390.596 560.883 560.823 500.606 65
SALANet0.670 560.816 360.770 480.768 550.652 520.807 650.451 610.747 610.659 250.545 600.924 940.473 570.149 1020.571 650.811 480.635 750.746 600.623 440.892 500.794 690.570 78
O3DSeg0.668 570.822 340.771 470.496 1060.651 530.833 390.541 190.761 560.555 700.611 380.966 140.489 500.370 60.388 1000.580 820.776 140.751 570.570 660.956 60.817 550.646 52
PointASNLpermissive0.666 580.703 820.781 420.751 620.655 500.830 420.471 520.769 540.474 910.537 630.951 460.475 560.279 450.635 460.698 690.675 560.751 570.553 770.816 890.806 600.703 35
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PointConvpermissive0.666 580.781 460.759 530.699 710.644 570.822 510.475 500.779 510.564 670.504 770.953 400.428 780.203 810.586 600.754 550.661 620.753 560.588 620.902 420.813 580.642 53
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PPCNN++permissive0.663 600.746 640.708 750.722 640.638 590.820 540.451 610.566 970.599 500.541 610.950 500.510 440.313 260.648 430.819 460.616 800.682 830.590 600.869 740.810 590.656 48
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
DCM-Net0.658 610.778 470.702 780.806 400.619 630.813 640.468 540.693 770.494 840.524 690.941 760.449 700.298 330.510 820.821 440.675 560.727 670.568 690.826 870.803 630.637 55
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
MVF-GNN0.658 610.558 1030.751 590.655 860.690 390.722 950.453 600.867 190.579 600.576 530.893 1060.523 390.293 350.733 310.571 840.692 480.659 900.606 510.875 640.804 620.668 44
HPGCNN0.656 630.698 840.743 660.650 880.564 800.820 540.505 360.758 570.631 340.479 810.945 640.480 540.226 680.572 640.774 530.690 510.735 630.614 470.853 810.776 840.597 71
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
SAFNet-segpermissive0.654 640.752 610.734 700.664 840.583 750.815 600.399 850.754 590.639 310.535 650.942 740.470 580.309 280.665 390.539 860.650 650.708 730.635 380.857 800.793 710.642 53
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Netpermissive0.645 650.778 470.731 710.699 710.577 760.829 430.446 660.736 650.477 900.523 710.945 640.454 650.269 540.484 900.749 580.618 780.738 610.599 550.827 860.792 740.621 60
PointConv-SFPN0.641 660.776 490.703 770.721 650.557 830.826 460.451 610.672 820.563 680.483 800.943 730.425 810.162 970.644 440.726 600.659 630.709 720.572 650.875 640.786 790.559 84
MVPNetpermissive0.641 660.831 300.715 730.671 810.590 710.781 790.394 870.679 790.642 290.553 580.937 810.462 610.256 600.649 420.406 1000.626 760.691 800.666 290.877 620.792 740.608 64
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
PointMRNet0.640 680.717 790.701 790.692 740.576 770.801 690.467 560.716 700.563 680.459 870.953 400.429 770.169 940.581 610.854 330.605 810.710 700.550 790.894 490.793 710.575 76
FPConvpermissive0.639 690.785 440.760 520.713 690.603 660.798 710.392 880.534 1020.603 480.524 690.948 560.457 630.250 620.538 760.723 620.598 850.696 780.614 470.872 700.799 640.567 81
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
PD-Net0.638 700.797 400.769 490.641 940.590 710.820 540.461 580.537 1010.637 320.536 640.947 580.388 910.206 780.656 400.668 730.647 690.732 650.585 630.868 750.793 710.473 104
PointSPNet0.637 710.734 700.692 860.714 680.576 770.797 720.446 660.743 630.598 510.437 920.942 740.403 870.150 1010.626 500.800 510.649 660.697 770.557 750.846 830.777 830.563 82
SConv0.636 720.830 310.697 820.752 610.572 790.780 810.445 680.716 700.529 740.530 660.951 460.446 720.170 930.507 850.666 740.636 740.682 830.541 850.886 540.799 640.594 72
Supervoxel-CNN0.635 730.656 890.711 740.719 660.613 640.757 900.444 710.765 550.534 730.566 550.928 920.478 550.272 500.636 450.531 880.664 600.645 940.508 920.864 770.792 740.611 61
joint point-basedpermissive0.634 740.614 970.778 430.667 830.633 610.825 470.420 790.804 450.467 930.561 560.951 460.494 470.291 370.566 660.458 950.579 910.764 470.559 740.838 840.814 560.598 70
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
PointMTL0.632 750.731 720.688 890.675 780.591 700.784 780.444 710.565 980.610 420.492 780.949 540.456 640.254 610.587 580.706 650.599 840.665 890.612 500.868 750.791 770.579 75
3DSM_DMMF0.631 760.626 940.745 640.801 430.607 650.751 910.506 350.729 680.565 660.491 790.866 1090.434 730.197 850.595 560.630 780.709 410.705 750.560 720.875 640.740 940.491 99
PointNet2-SFPN0.631 760.771 530.692 860.672 790.524 880.837 340.440 730.706 750.538 720.446 890.944 700.421 830.219 730.552 720.751 570.591 870.737 620.543 840.901 440.768 860.557 85
APCF-Net0.631 760.742 670.687 910.672 790.557 830.792 760.408 810.665 830.545 710.508 740.952 440.428 780.186 880.634 470.702 670.620 770.706 740.555 760.873 680.798 660.581 74
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
FusionAwareConv0.630 790.604 990.741 680.766 570.590 710.747 920.501 380.734 660.503 830.527 670.919 980.454 650.323 230.550 740.420 990.678 550.688 810.544 820.896 470.795 680.627 59
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DenSeR0.628 800.800 390.625 1020.719 660.545 850.806 660.445 680.597 910.448 980.519 720.938 800.481 530.328 210.489 890.499 930.657 640.759 520.592 590.881 580.797 670.634 56
SegGroup_sempermissive0.627 810.818 350.747 630.701 700.602 670.764 870.385 920.629 880.490 860.508 740.931 910.409 860.201 820.564 670.725 610.618 780.692 790.539 860.873 680.794 690.548 88
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
SIConv0.625 820.830 310.694 840.757 590.563 810.772 850.448 650.647 860.520 770.509 730.949 540.431 760.191 860.496 870.614 800.647 690.672 870.535 880.876 630.783 800.571 77
dtc_net0.625 820.703 820.751 590.794 460.535 860.848 240.480 490.676 810.528 750.469 840.944 700.454 650.004 1150.464 920.636 770.704 440.758 530.548 810.924 270.787 780.492 98
HPEIN0.618 840.729 730.668 920.647 900.597 690.766 860.414 800.680 780.520 770.525 680.946 610.432 740.215 750.493 880.599 810.638 730.617 990.570 660.897 460.806 600.605 67
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
SPH3D-GCNpermissive0.610 850.858 240.772 450.489 1070.532 870.792 760.404 840.643 870.570 650.507 760.935 840.414 850.046 1120.510 820.702 670.602 830.705 750.549 800.859 790.773 850.534 91
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
AttAN0.609 860.760 580.667 930.649 890.521 890.793 740.457 590.648 850.528 750.434 940.947 580.401 880.153 1000.454 930.721 630.648 680.717 690.536 870.904 390.765 870.485 100
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
wsss-transformer0.600 870.634 930.743 660.697 730.601 680.781 790.437 750.585 940.493 850.446 890.933 890.394 890.011 1140.654 410.661 760.603 820.733 640.526 890.832 850.761 890.480 101
LAP-D0.594 880.720 770.692 860.637 950.456 990.773 840.391 900.730 670.587 550.445 910.940 780.381 920.288 380.434 960.453 970.591 870.649 920.581 640.777 930.749 930.610 63
DPC0.592 890.720 770.700 800.602 990.480 950.762 890.380 930.713 730.585 580.437 920.940 780.369 940.288 380.434 960.509 920.590 890.639 970.567 700.772 950.755 910.592 73
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
CCRFNet0.589 900.766 570.659 970.683 760.470 980.740 940.387 910.620 900.490 860.476 820.922 960.355 970.245 650.511 810.511 910.571 920.643 950.493 960.872 700.762 880.600 69
ROSMRF0.580 910.772 520.707 760.681 770.563 810.764 870.362 950.515 1030.465 940.465 860.936 830.427 800.207 770.438 940.577 830.536 950.675 860.486 970.723 1010.779 810.524 94
SD-DETR0.576 920.746 640.609 1060.445 1110.517 900.643 1060.366 940.714 720.456 960.468 850.870 1080.432 740.264 570.558 700.674 710.586 900.688 810.482 980.739 990.733 960.537 90
SQN_0.1%0.569 930.676 860.696 830.657 850.497 910.779 820.424 770.548 990.515 790.376 990.902 1050.422 820.357 90.379 1010.456 960.596 860.659 900.544 820.685 1040.665 1070.556 86
TextureNetpermissive0.566 940.672 880.664 940.671 810.494 930.719 960.445 680.678 800.411 1040.396 970.935 840.356 960.225 700.412 980.535 870.565 930.636 980.464 1000.794 920.680 1040.568 80
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
DVVNet0.562 950.648 900.700 800.770 540.586 740.687 1000.333 990.650 840.514 800.475 830.906 1020.359 950.223 720.340 1030.442 980.422 1060.668 880.501 930.708 1020.779 810.534 91
Pointnet++ & Featurepermissive0.557 960.735 690.661 960.686 750.491 940.744 930.392 880.539 1000.451 970.375 1000.946 610.376 930.205 790.403 990.356 1030.553 940.643 950.497 940.824 880.756 900.515 95
GMLPs0.538 970.495 1080.693 850.647 900.471 970.793 740.300 1020.477 1040.505 820.358 1020.903 1040.327 1000.081 1090.472 910.529 890.448 1040.710 700.509 900.746 970.737 950.554 87
PanopticFusion-label0.529 980.491 1090.688 890.604 980.386 1040.632 1070.225 1120.705 760.434 1010.293 1080.815 1100.348 980.241 660.499 860.669 720.507 970.649 920.442 1060.796 910.602 1110.561 83
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
subcloud_weak0.516 990.676 860.591 1090.609 960.442 1000.774 830.335 980.597 910.422 1030.357 1030.932 900.341 990.094 1080.298 1050.528 900.473 1020.676 850.495 950.602 1100.721 990.349 111
Online SegFusion0.515 1000.607 980.644 1000.579 1010.434 1010.630 1080.353 960.628 890.440 990.410 950.762 1140.307 1020.167 950.520 790.403 1010.516 960.565 1020.447 1040.678 1050.701 1010.514 96
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
3DMV, FTSDF0.501 1010.558 1030.608 1070.424 1130.478 960.690 990.246 1080.586 930.468 920.450 880.911 1000.394 890.160 980.438 940.212 1100.432 1050.541 1080.475 990.742 980.727 970.477 102
PCNN0.498 1020.559 1020.644 1000.560 1030.420 1030.711 980.229 1100.414 1050.436 1000.352 1040.941 760.324 1010.155 990.238 1100.387 1020.493 980.529 1090.509 900.813 900.751 920.504 97
Weakly-Openseg v30.489 1030.749 630.664 940.646 920.496 920.559 1120.122 1150.577 950.257 1150.364 1010.805 1110.198 1130.096 1070.510 820.496 940.361 1100.563 1030.359 1130.777 930.644 1080.532 93
3DMV0.484 1040.484 1100.538 1110.643 930.424 1020.606 1110.310 1000.574 960.433 1020.378 980.796 1120.301 1030.214 760.537 770.208 1110.472 1030.507 1120.413 1090.693 1030.602 1110.539 89
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PointCNN with RGBpermissive0.458 1050.577 1010.611 1050.356 1150.321 1120.715 970.299 1040.376 1090.328 1110.319 1060.944 700.285 1050.164 960.216 1130.229 1080.484 1000.545 1070.456 1020.755 960.709 1000.475 103
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
FCPNpermissive0.447 1060.679 850.604 1080.578 1020.380 1050.682 1010.291 1050.106 1150.483 890.258 1130.920 970.258 1090.025 1130.231 1120.325 1040.480 1010.560 1050.463 1010.725 1000.666 1060.231 115
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
DGCNN_reproducecopyleft0.446 1070.474 1110.623 1030.463 1090.366 1070.651 1040.310 1000.389 1080.349 1090.330 1050.937 810.271 1070.126 1040.285 1060.224 1090.350 1120.577 1010.445 1050.625 1080.723 980.394 107
Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon: Dynamic Graph CNN for Learning on Point Clouds. TOG 2019
PNET20.442 1080.548 1050.548 1100.597 1000.363 1080.628 1090.300 1020.292 1100.374 1060.307 1070.881 1070.268 1080.186 880.238 1100.204 1120.407 1070.506 1130.449 1030.667 1060.620 1100.462 105
SurfaceConvPF0.442 1080.505 1070.622 1040.380 1140.342 1100.654 1030.227 1110.397 1070.367 1070.276 1100.924 940.240 1100.198 840.359 1020.262 1060.366 1080.581 1000.435 1070.640 1070.668 1050.398 106
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
Tangent Convolutionspermissive0.438 1100.437 1130.646 990.474 1080.369 1060.645 1050.353 960.258 1120.282 1130.279 1090.918 990.298 1040.147 1030.283 1070.294 1050.487 990.562 1040.427 1080.619 1090.633 1090.352 110
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
3DWSSS0.425 1110.525 1060.647 980.522 1040.324 1110.488 1150.077 1160.712 740.353 1080.401 960.636 1160.281 1060.176 910.340 1030.565 850.175 1160.551 1060.398 1100.370 1160.602 1110.361 109
SPLAT Netcopyleft0.393 1120.472 1120.511 1120.606 970.311 1130.656 1020.245 1090.405 1060.328 1110.197 1140.927 930.227 1120.000 1170.001 1170.249 1070.271 1150.510 1100.383 1120.593 1110.699 1020.267 113
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+FTSDF0.383 1130.297 1150.491 1130.432 1120.358 1090.612 1100.274 1060.116 1140.411 1040.265 1110.904 1030.229 1110.079 1100.250 1080.185 1130.320 1130.510 1100.385 1110.548 1120.597 1140.394 107
PointNet++permissive0.339 1140.584 1000.478 1140.458 1100.256 1150.360 1160.250 1070.247 1130.278 1140.261 1120.677 1150.183 1140.117 1050.212 1140.145 1150.364 1090.346 1160.232 1160.548 1120.523 1150.252 114
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNetpermissive0.308 1150.353 1140.290 1160.278 1160.166 1160.553 1130.169 1140.286 1110.147 1160.148 1160.908 1010.182 1150.064 1110.023 1160.018 1170.354 1110.363 1140.345 1140.546 1140.685 1030.278 112
ScanNetpermissive0.306 1160.203 1160.366 1150.501 1050.311 1130.524 1140.211 1130.002 1170.342 1100.189 1150.786 1130.145 1160.102 1060.245 1090.152 1140.318 1140.348 1150.300 1150.460 1150.437 1160.182 116
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
ERROR0.054 1170.000 1170.041 1170.172 1170.030 1170.062 1170.001 1170.035 1160.004 1170.051 1170.143 1170.019 1170.003 1160.041 1150.050 1160.003 1170.054 1170.018 1170.005 1170.264 1170.082 117


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




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
PointRel0.622 10.926 80.710 20.541 80.502 20.772 40.314 40.598 110.425 70.504 70.565 10.650 50.716 20.809 70.476 100.747 40.618 10.963 30.364 18
: Relation3D (PointRel): Enhancing Relation Modeling for Point Cloud Instance Segmentation.
SIM3D0.617 20.952 40.629 140.539 90.426 130.768 80.302 60.681 20.425 80.473 130.511 130.701 10.717 10.821 60.467 130.774 10.559 130.914 150.448 2
Spherical Mask(CtoF)0.616 30.946 50.654 100.555 50.434 100.769 70.271 100.604 80.447 40.505 50.549 20.698 20.716 20.775 140.480 70.747 50.575 90.925 110.436 4
EV3D0.615 40.946 50.652 110.555 50.433 110.773 30.271 110.604 80.447 40.506 40.544 50.698 20.716 20.775 140.480 70.747 50.572 110.925 110.435 5
ExtMask3D0.598 50.852 150.692 60.433 280.461 60.791 10.264 120.488 330.493 10.508 30.528 120.594 100.706 60.791 80.483 50.734 90.595 30.911 170.437 3
MAFT0.596 60.889 130.721 10.448 210.460 70.768 90.251 130.558 210.408 90.504 60.539 70.616 80.618 100.858 30.482 60.684 180.551 150.931 100.450 1
UniPerception0.588 70.963 30.667 80.493 130.472 50.750 120.229 160.528 260.468 30.498 100.542 60.643 60.530 190.661 350.463 140.695 170.599 20.972 10.420 6
MG-Former0.587 80.852 150.639 130.454 200.393 180.758 110.338 20.572 160.480 20.527 20.491 190.671 40.527 200.867 10.485 40.601 280.590 60.938 90.390 10
InsSSM0.586 91.000 10.593 180.440 240.480 30.771 50.345 10.437 370.444 60.495 110.548 40.579 130.621 90.720 260.409 200.712 110.593 40.960 40.395 8
Queryformer0.583 100.926 80.702 40.393 340.504 10.733 180.276 90.527 270.373 150.479 120.534 90.533 200.697 70.720 270.436 180.745 70.592 50.958 50.363 19
Competitor-SPFormer0.580 110.721 310.705 30.593 30.444 90.786 20.286 70.564 190.376 140.498 90.534 100.546 180.390 410.785 100.577 10.708 150.579 80.954 60.388 11
PBNetpermissive0.573 120.926 80.575 230.619 10.472 40.736 160.239 150.487 340.383 130.459 160.506 160.533 190.585 120.767 160.404 210.717 100.559 140.969 20.381 14
W.Zhao, Y.Yan, C.Yang, J.Ye,X.Yang,K.Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
TST3D0.569 130.778 230.675 70.598 20.451 80.727 190.280 80.476 360.395 100.472 140.457 250.583 110.580 140.777 110.462 160.735 80.547 170.919 140.333 25
Duc Tran Dang Trung, Byeongkeun Kang, Yeejin Lee: MSTA3D: Multi-scale Twin-attention for 3D Instance Segmentation. ACM Multimedia 2024
Mask3D0.566 140.926 80.597 170.408 310.420 150.737 150.239 140.598 110.386 120.458 170.549 20.568 160.716 20.601 410.480 70.646 220.575 90.922 130.364 17
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
OneFormer3Dcopyleft0.566 140.781 220.697 50.562 40.431 120.770 60.331 30.400 430.373 160.529 10.504 170.568 150.475 250.732 240.470 110.762 20.550 160.871 320.379 15
Maxim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: OneFormer3D: One Transformer for Unified Point Cloud Segmentation.
ISBNetpermissive0.559 160.939 70.655 90.383 370.426 140.763 100.180 180.534 250.386 110.499 80.509 150.621 70.427 350.704 300.467 120.649 210.571 120.948 70.401 7
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
GraphCut0.552 171.000 10.611 160.438 250.392 190.714 200.139 210.598 130.327 190.389 200.510 140.598 90.427 360.754 190.463 150.761 30.588 70.903 200.329 26
SPFormerpermissive0.549 180.745 260.640 120.484 140.395 170.739 140.311 50.566 180.335 180.468 150.492 180.555 170.478 240.747 210.436 170.712 120.540 180.893 240.343 24
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
DKNet0.532 190.815 190.624 150.517 100.377 210.749 130.107 230.509 300.304 210.437 180.475 200.581 120.539 170.775 130.339 260.640 240.506 210.901 210.385 13
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
IPCA-Inst0.520 200.889 130.551 270.548 70.418 160.665 300.064 320.585 140.260 290.277 340.471 220.500 210.644 80.785 90.369 220.591 310.511 190.878 290.362 20
SoftGroup++0.513 210.704 330.578 220.398 330.363 270.704 210.061 330.647 50.297 260.378 230.537 80.343 240.614 110.828 50.295 310.710 140.505 230.875 310.394 9
SSTNetpermissive0.506 220.738 290.549 280.497 120.316 320.693 240.178 190.377 460.198 350.330 250.463 240.576 140.515 210.857 40.494 20.637 250.457 270.943 80.290 35
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
SoftGrouppermissive0.504 230.667 400.579 200.372 390.381 200.694 230.072 290.677 30.303 220.387 210.531 110.319 280.582 130.754 180.318 270.643 230.492 240.907 190.388 12
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
DANCENET0.504 230.926 80.579 190.472 160.367 240.626 400.165 200.432 380.221 310.408 190.449 270.411 220.564 150.746 220.421 190.707 160.438 300.846 400.288 36
TD3Dpermissive0.489 250.852 150.511 370.434 260.322 310.735 170.101 260.512 290.355 170.349 240.468 230.283 320.514 220.676 340.268 360.671 190.510 200.908 180.329 27
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
OccuSeg+instance0.486 260.802 210.536 300.428 290.369 230.702 220.205 170.331 510.301 230.379 220.474 210.327 250.437 300.862 20.485 30.601 290.394 380.846 420.273 39
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
TopoSeg0.479 270.704 330.564 240.467 180.366 250.633 380.068 300.554 220.262 280.328 260.447 280.323 260.534 180.722 250.288 330.614 260.482 250.912 160.358 22
DualGroup0.469 280.815 190.552 260.398 320.374 220.683 260.130 220.539 240.310 200.327 270.407 310.276 330.447 290.535 450.342 250.659 200.455 280.900 230.301 31
SSEC0.465 290.667 400.578 210.502 110.362 280.641 370.035 420.605 70.291 270.323 280.451 260.296 300.417 390.677 330.245 400.501 490.506 220.900 220.366 16
HAISpermissive0.457 300.704 330.561 250.457 190.364 260.673 270.046 410.547 230.194 360.308 290.426 290.288 310.454 280.711 280.262 370.563 390.434 320.889 260.344 23
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
DD-UNet+Group0.436 310.630 480.508 400.480 150.310 340.624 420.065 310.638 60.174 370.256 380.384 350.194 450.428 330.759 170.289 320.574 360.400 360.849 390.291 34
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
INS-Conv-instance0.435 320.716 320.495 420.355 410.331 290.689 250.102 250.394 450.208 340.280 320.395 330.250 360.544 160.741 230.309 290.536 450.391 390.842 450.258 43
Mask-Group0.434 330.778 230.516 350.471 170.330 300.658 310.029 440.526 280.249 300.256 370.400 320.309 290.384 440.296 610.368 230.575 350.425 330.877 300.362 21
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
Box2Mask0.433 340.741 270.463 470.433 270.283 370.625 410.103 240.298 560.125 460.260 360.424 300.322 270.472 260.701 310.363 240.711 130.309 550.882 270.272 41
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
RPGN0.428 350.630 480.508 390.367 400.249 440.658 320.016 520.673 40.131 440.234 410.383 360.270 340.434 310.748 200.274 350.609 270.406 350.842 440.267 42
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
DENet0.413 360.741 270.520 320.237 520.284 360.523 510.097 270.691 10.138 410.209 510.229 530.238 390.390 420.707 290.310 280.448 560.470 260.892 250.310 29
PointGroup0.407 370.639 470.496 410.415 300.243 460.645 360.021 490.570 170.114 470.211 490.359 380.217 430.428 340.660 360.256 380.562 400.341 470.860 350.291 33
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]
CSC-Pretrained0.405 380.738 290.465 460.331 450.205 500.655 330.051 370.601 100.092 510.211 500.329 410.198 440.459 270.775 120.195 470.524 470.400 370.878 280.184 52
PE0.396 390.667 400.467 450.446 230.243 450.624 430.022 480.577 150.106 480.219 440.340 390.239 380.487 230.475 520.225 420.541 440.350 450.818 470.273 40
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
Dyco3Dcopyleft0.395 400.642 460.518 340.447 220.259 430.666 290.050 380.251 610.166 380.231 420.362 370.232 400.331 470.535 440.229 410.587 320.438 310.850 370.317 28
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
OSIS0.392 410.778 230.530 310.220 540.278 380.567 480.083 280.330 520.299 240.270 350.310 440.143 510.260 510.624 390.277 340.568 380.361 430.865 340.301 30
AOIA0.387 420.704 330.515 360.385 360.225 490.669 280.005 590.482 350.126 450.181 540.269 500.221 420.426 370.478 510.218 430.592 300.371 410.851 360.242 45
SSEN0.384 430.852 150.494 430.192 550.226 480.648 350.022 470.398 440.299 250.277 330.317 430.231 410.194 580.514 480.196 450.586 330.444 290.843 430.184 51
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
Mask3D_evaluation0.382 440.593 500.520 330.390 350.314 330.600 440.018 510.287 590.151 400.281 310.387 340.169 490.429 320.654 370.172 510.578 340.384 400.670 580.278 38
PCJC0.375 450.704 330.542 290.284 490.197 520.649 340.006 560.426 390.138 420.242 390.304 450.183 480.388 430.629 380.141 580.546 430.344 460.738 530.283 37
ClickSeg_Instance0.366 460.654 440.375 510.184 560.302 350.592 460.050 390.300 550.093 500.283 300.277 470.249 370.426 380.615 400.299 300.504 480.367 420.832 460.191 50
SphereSeg0.357 470.651 450.411 490.345 420.264 420.630 390.059 340.289 580.212 320.240 400.336 400.158 500.305 480.557 420.159 540.455 550.341 480.726 550.294 32
3D-MPA0.355 480.457 600.484 440.299 470.277 390.591 470.047 400.332 490.212 330.217 450.278 460.193 460.413 400.410 550.195 460.574 370.352 440.849 380.213 48
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
NeuralBF