The 3D semantic labeling task involves predicting a semantic labeling of a 3D scan mesh.

Evaluation and metrics

Our evaluation ranks all methods according to the PASCAL VOC intersection-over-union metric (IoU). IoU = TP/(TP+FP+FN), where TP, FP, and FN are the numbers of true positive, false positive, and false negative pixels, respectively. Predicted labels are evaluated per-vertex over the respective 3D scan mesh; for 3D approaches that operate on other representations like grids or points, the predicted labels should be mapped onto the mesh vertices (e.g., one such example for grid to mesh vertices is provided in the evaluation helpers).



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 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 bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort 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
ALS-MinkowskiNetcopyleft0.414 30.610 30.322 30.271 20.542 20.153 30.159 120.000 30.000 80.000 10.404 40.503 50.532 70.672 170.804 50.285 10.888 30.000 30.900 30.226 30.087 20.598 50.342 50.671 10.217 110.087 40.449 40.000 10.000 30.253 30.477 71.000 10.000 10.118 60.000 30.905 10.071 140.710 30.076 30.047 170.665 20.376 90.981 10.000 10.000 20.466 70.632 80.113 40.769 10.956 50.795 20.031 90.314 10.936 10.000 10.390 20.601 40.000 70.458 90.366 30.719 40.440 60.564 10.699 40.314 10.464 80.784 30.200 10.283 60.973 10.142 90.000 10.250 80.285 60.220 80.718 10.752 60.723 20.460 10.248 160.475 100.463 140.000 40.000 10.446 90.021 50.025 110.285 10.000 50.972 10.149 80.769 10.230 30.535 10.879 30.252 90.000 10.693 10.129 20.000 140.000 40.000 10.447 10.958 10.662 90.159 20.598 40.780 120.344 20.646 40.106 60.893 30.135 30.455 40.000 10.194 30.259 10.726 30.475 40.000 90.000 10.741 10.865 20.571 20.817 30.445 40.000 10.506 30.630 40.230 130.916 20.728 10.635 11.000 10.252 70.000 10.804 30.697 80.137 110.043 80.717 30.807 40.000 10.510 140.245 20.000 70.000 10.709 30.000 20.000 10.703 30.572 50.646 20.223 110.531 60.984 10.397 40.813 10.798 10.135 130.800 10.000 10.097 20.832 30.752 90.842 80.000 10.852 10.149 100.846 110.000 10.666 50.359 60.252 90.777 10.690 2
Guangda Ji, Silvan Weder, Francis Engelmann, Marc Pollefeys, Hermann Blum: ARKit LabelMaker: A New Scale for Indoor 3D Scene Understanding. CVPR 2025
AWCS0.305 150.508 150.225 150.142 120.463 140.063 140.195 100.000 30.000 80.000 10.467 30.551 30.504 90.773 70.764 150.142 140.029 180.000 30.626 140.100 120.000 100.360 140.179 160.507 140.137 160.006 90.300 130.000 10.000 30.172 90.364 160.512 100.000 10.056 150.000 30.865 140.093 40.634 180.000 80.071 140.396 150.296 170.876 100.000 10.000 20.373 140.436 170.063 90.749 20.877 110.721 130.131 30.124 150.804 160.000 10.000 110.515 130.010 60.452 110.252 100.578 150.417 90.179 180.484 100.171 70.337 150.606 130.000 70.115 110.937 150.142 90.000 10.008 120.000 150.157 170.484 150.402 180.501 160.339 100.553 70.529 30.478 130.000 40.000 10.404 110.001 110.022 130.077 100.000 50.894 130.219 70.628 70.093 160.305 150.886 10.233 100.000 10.603 100.112 60.023 90.000 40.000 10.000 120.741 70.664 80.097 160.253 150.782 110.264 120.523 120.154 20.707 170.000 60.411 90.000 10.000 80.000 30.332 170.000 50.000 90.000 10.602 80.595 110.185 140.656 170.159 70.000 10.355 120.424 160.154 160.729 160.516 110.220 110.620 40.084 140.000 10.707 150.651 140.173 50.014 100.381 180.582 150.000 10.619 30.049 130.000 70.000 10.702 40.000 20.000 10.302 170.489 160.317 140.334 70.392 80.922 150.254 140.533 140.394 140.129 150.613 160.000 10.000 120.820 60.649 120.749 140.000 10.782 150.282 60.863 70.000 10.288 160.006 130.220 120.633 150.542 3
: Long-Tailed 3D Semantic Segmentation with Adaptive Weight Constraint and Sampling. ICRA 2024
DITR0.449 10.629 10.392 10.289 10.650 10.168 20.862 10.000 30.313 40.000 10.580 10.568 20.564 40.766 80.867 10.238 50.949 10.000 30.866 40.300 10.000 100.664 10.482 10.508 130.317 10.420 10.551 20.000 10.000 30.486 20.519 10.662 50.000 10.385 10.000 30.901 30.079 100.727 20.000 80.160 30.606 40.417 50.967 30.000 10.000 20.498 50.596 120.130 20.728 30.998 10.805 10.000 170.314 10.934 20.000 10.278 40.636 10.000 70.403 130.367 20.741 30.484 20.500 21.000 10.113 120.828 10.815 20.000 70.733 20.969 40.374 20.000 10.579 11.000 10.230 60.617 60.983 10.729 10.423 40.855 10.508 60.622 20.018 30.000 10.591 30.034 40.028 100.066 120.869 10.904 80.334 20.651 50.716 10.514 20.871 70.315 40.000 10.664 30.128 30.014 100.000 40.000 10.392 30.851 30.817 10.153 30.823 10.991 10.318 40.680 20.134 30.913 10.157 20.448 50.000 10.000 80.000 30.826 10.978 10.091 60.000 10.660 50.647 40.571 20.804 40.001 100.000 10.480 40.700 10.421 60.947 10.433 150.411 40.148 70.262 50.000 10.849 10.709 70.138 100.150 20.714 40.889 20.000 10.698 10.222 40.000 70.000 10.720 20.000 20.000 10.805 10.600 20.642 30.268 100.904 10.982 20.477 20.632 70.718 20.139 100.776 20.000 10.178 10.886 20.962 10.839 90.000 10.851 20.043 130.869 50.000 10.710 10.315 70.348 40.753 20.397 8
Karim Abou Zeid, Kadir Yilmaz, Daan de Geus, Alexander Hermans, David Adrian, Timm Linder, Bastian Leibe: DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation. 3DV 2026
PPT-SpUNet-F.T.0.332 130.556 70.270 80.123 150.519 50.091 80.349 50.000 30.000 80.000 10.339 100.383 110.498 110.833 50.807 40.241 40.584 100.000 30.755 80.124 90.000 100.608 30.330 80.530 100.314 20.000 100.374 90.000 10.000 30.197 60.459 80.000 120.000 10.117 70.000 30.876 80.095 20.682 100.000 80.086 90.518 80.433 30.930 50.000 10.000 20.563 30.542 150.077 70.715 40.858 120.756 60.008 160.171 130.874 90.000 10.039 70.550 120.000 70.545 50.256 90.657 90.453 50.351 110.449 110.213 60.392 130.611 120.000 70.037 160.946 70.138 140.000 10.000 130.063 110.308 20.537 90.796 50.673 50.323 120.392 110.400 150.509 80.000 40.000 10.649 10.000 120.023 120.000 130.000 50.914 70.002 170.506 170.163 120.359 90.872 60.000 130.000 10.623 80.112 60.001 120.000 40.000 10.021 100.753 60.565 160.150 50.579 50.806 100.267 100.616 50.042 150.783 140.000 60.374 120.000 10.000 80.000 30.620 90.000 50.000 90.000 10.572 140.634 60.350 100.792 50.000 110.000 10.376 100.535 70.378 70.855 80.672 40.074 140.000 120.185 110.000 10.727 130.660 130.076 180.000 110.432 130.646 110.000 10.594 80.006 140.000 70.000 10.658 80.000 20.000 10.661 50.549 110.300 150.291 90.045 150.942 120.304 90.600 90.572 80.135 130.695 60.000 10.008 90.793 100.942 20.899 30.000 10.816 70.181 80.897 30.000 10.679 40.223 90.264 60.691 60.345 12
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
PTv3 ScanNet2000.393 40.592 40.330 20.216 40.520 40.109 60.108 170.000 30.337 20.000 10.310 130.394 100.494 120.753 100.848 20.256 30.717 90.000 30.842 50.192 60.065 40.449 110.346 40.546 70.190 140.000 100.384 80.000 10.000 30.218 50.505 20.791 30.000 10.136 50.000 30.903 20.073 130.687 70.000 80.168 20.551 60.387 80.941 40.000 10.000 20.397 130.654 30.000 100.714 50.759 160.752 80.118 40.264 50.926 30.000 10.048 60.575 60.000 70.597 20.366 30.755 10.469 30.474 30.798 20.140 100.617 30.692 80.000 70.592 40.971 20.188 40.000 10.133 100.593 20.349 10.650 40.717 90.699 40.455 20.790 20.523 40.636 10.301 10.000 10.622 20.000 120.017 150.259 30.000 50.921 40.337 10.733 20.210 50.514 20.860 90.407 10.000 10.688 20.109 80.000 140.000 40.000 10.151 60.671 90.782 20.115 140.641 20.903 20.349 10.616 50.088 70.832 90.000 60.480 30.000 10.428 10.000 30.497 110.000 50.000 90.000 10.662 40.690 30.612 10.828 10.575 20.000 10.404 80.644 20.325 80.887 50.728 10.009 170.134 80.026 180.000 10.761 40.731 50.172 60.077 40.528 90.727 80.000 10.603 50.220 50.022 30.000 10.740 10.000 20.000 10.661 50.586 30.566 50.436 40.531 60.978 40.457 30.708 40.583 70.141 70.748 30.000 10.026 50.822 40.871 40.879 60.000 10.851 20.405 20.914 10.000 10.682 30.000 160.281 50.738 30.463 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)
LGroundpermissive0.272 160.485 160.184 160.106 160.476 120.077 110.218 90.000 30.000 80.000 10.547 20.295 120.540 60.746 110.745 160.058 170.112 170.005 10.658 120.077 160.000 100.322 150.178 170.512 120.190 140.199 20.277 160.000 10.000 30.173 80.399 130.000 120.000 10.039 170.000 30.858 150.085 70.676 120.002 60.103 60.498 90.323 150.703 130.000 10.000 20.296 160.549 130.216 10.702 60.768 150.718 150.028 100.092 170.786 170.000 10.000 110.453 170.022 50.251 180.252 100.572 160.348 150.321 120.514 70.063 150.279 170.552 160.000 70.019 170.932 160.132 160.000 10.000 130.000 150.156 180.457 160.623 130.518 150.265 170.358 120.381 160.395 150.000 40.000 10.127 180.012 90.051 10.000 130.000 50.886 140.014 140.437 180.179 90.244 160.826 160.000 130.000 10.599 110.136 10.085 30.000 40.000 10.000 120.565 140.612 140.143 60.207 160.566 150.232 160.446 160.127 40.708 160.000 60.384 100.000 10.000 80.000 30.402 150.000 50.059 70.000 10.525 160.566 120.229 130.659 160.000 110.000 10.265 160.446 150.147 170.720 180.597 90.066 150.000 120.187 100.000 10.726 140.467 180.134 130.000 110.413 160.629 130.000 10.363 170.055 110.022 30.000 10.626 120.000 20.000 10.323 160.479 180.154 170.117 160.028 170.901 160.243 160.415 170.295 180.143 60.610 170.000 10.000 120.777 130.397 180.324 170.000 10.778 160.179 90.702 170.000 10.274 170.404 50.233 110.622 160.398 7
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
L3DETR-ScanNet_2000.336 90.533 120.279 70.155 110.508 60.073 120.101 180.000 30.058 70.000 10.294 150.233 150.548 50.927 20.788 110.264 20.463 120.000 30.638 130.098 140.014 80.411 130.226 140.525 110.225 100.010 80.397 70.000 10.000 30.192 70.380 150.598 70.000 10.117 70.000 30.883 70.082 80.689 50.000 80.032 180.549 70.417 50.910 60.000 10.000 20.448 80.613 100.000 100.697 70.960 40.759 50.158 20.293 30.883 80.000 10.312 30.583 50.079 40.422 120.068 180.660 80.418 80.298 130.430 130.114 110.526 60.776 40.051 30.679 30.946 70.152 70.000 10.183 90.000 150.211 90.511 110.409 170.565 130.355 90.448 80.512 50.557 40.000 40.000 10.420 100.000 120.007 170.104 70.000 50.125 180.330 30.514 160.146 130.321 140.860 90.174 120.000 10.629 70.075 140.000 140.000 40.000 10.002 110.671 90.712 70.141 70.339 130.856 50.261 130.529 110.067 100.835 70.000 60.369 130.000 10.259 20.000 30.629 70.000 50.487 10.000 10.579 120.646 50.107 180.720 110.122 80.000 10.333 150.505 110.303 100.908 40.503 140.565 20.074 90.324 10.000 10.740 90.661 120.109 140.000 110.427 140.563 180.000 10.579 110.108 80.000 70.000 10.664 70.000 20.000 10.641 80.539 120.416 80.515 20.256 120.940 130.312 70.209 180.620 40.138 120.636 120.000 10.000 120.775 140.861 50.765 130.000 10.801 100.119 120.860 90.000 10.687 20.001 150.192 150.679 100.699 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12
Voltpermissive0.416 20.619 20.318 40.269 30.528 30.138 40.862 10.000 30.356 10.000 10.380 80.438 70.616 20.952 10.795 70.143 130.891 20.000 30.904 20.227 20.087 20.606 40.237 130.625 20.238 80.188 30.429 50.000 10.000 30.251 40.504 30.791 30.000 10.218 40.000 30.900 50.082 80.735 10.097 10.093 80.754 10.475 10.981 10.000 10.000 20.425 90.653 40.000 100.696 80.988 20.773 30.000 170.265 40.905 50.000 10.000 110.631 20.000 70.493 80.401 10.753 20.499 10.392 90.437 120.000 170.609 40.881 10.000 70.277 70.958 50.142 90.000 10.518 20.000 150.274 40.700 20.752 60.709 30.421 50.431 90.462 110.583 30.000 40.000 10.553 50.020 60.007 170.218 40.631 20.934 20.005 160.614 80.223 40.430 40.884 20.407 10.000 10.652 50.040 180.000 140.000 40.000 10.398 20.855 20.635 110.151 40.624 30.903 20.335 30.686 10.063 110.865 40.000 60.551 10.000 10.000 80.000 30.678 40.000 50.000 90.000 10.696 20.962 10.410 80.679 150.997 10.000 10.542 20.635 30.588 10.909 30.728 10.414 31.000 10.261 60.000 10.834 20.737 40.136 120.066 50.888 10.924 10.000 10.541 120.069 100.000 70.000 10.682 60.000 20.000 10.747 20.639 10.603 40.329 80.778 20.982 20.501 10.725 30.680 30.141 70.719 40.000 10.000 120.893 10.842 60.930 10.000 10.850 40.272 70.898 20.000 10.351 140.576 10.357 30.721 40.324 13
Kadir Yilmaz, Adrian Kruse, Tristan Höfer, Daan de Geus, Bastian Leibe: Volume Transformer: Revisiting Vanilla Transformers for 3D Scene Understanding.
CeCo0.340 80.551 100.247 140.181 70.475 130.057 160.142 130.000 30.000 80.000 10.387 60.463 60.499 100.924 30.774 120.213 60.257 140.000 30.546 160.100 120.006 90.615 20.177 180.534 80.246 60.000 100.400 60.000 10.338 10.006 170.484 60.609 60.000 10.083 120.000 30.873 100.089 50.661 150.000 80.048 160.560 50.408 70.892 90.000 10.000 20.586 10.616 90.000 100.692 90.900 90.721 130.162 10.228 70.860 120.000 10.000 110.575 60.083 30.550 40.347 50.624 140.410 110.360 100.740 30.109 130.321 160.660 90.000 70.121 100.939 140.143 80.000 10.400 30.003 130.190 120.564 70.652 110.615 120.421 50.304 140.579 10.547 60.000 40.000 10.296 150.000 120.030 90.096 80.000 50.916 50.037 130.551 130.171 100.376 80.865 80.286 60.000 10.633 60.102 120.027 80.011 30.000 10.000 120.474 150.742 50.133 80.311 140.824 90.242 140.503 150.068 90.828 100.000 60.429 80.000 10.063 50.000 30.781 20.000 50.000 90.000 10.665 30.633 70.450 60.818 20.000 110.000 10.429 60.532 80.226 140.825 120.510 120.377 60.709 30.079 150.000 10.753 60.683 90.102 160.063 60.401 170.620 140.000 10.619 30.000 150.000 70.000 10.595 140.000 20.000 10.345 150.564 70.411 90.603 10.384 90.945 100.266 120.643 60.367 150.304 10.663 110.000 10.010 70.726 160.767 80.898 40.000 10.784 140.435 10.861 80.000 10.447 110.000 160.257 80.656 120.377 10
Zhisheng Zhong, Jiequan Cui, Yibo Yang, Xiaoyang Wu, Xiaojuan Qi, Xiangyu Zhang, Jiaya Jia: Understanding Imbalanced Semantic Segmentation Through Neural Collapse. CVPR 2023
OctFormer ScanNet200permissive0.326 140.539 110.265 110.131 130.499 70.110 50.522 40.000 30.000 80.000 10.318 120.427 80.455 160.743 120.765 140.175 110.842 50.000 30.828 60.204 50.033 70.429 120.335 60.601 30.312 30.000 100.357 110.000 10.000 30.047 120.423 100.000 120.000 10.105 100.000 30.873 100.079 100.670 130.000 80.117 50.471 140.432 40.829 120.000 10.000 20.584 20.417 180.089 60.684 100.837 130.705 170.021 120.178 120.892 70.000 10.028 80.505 140.000 70.457 100.200 150.662 50.412 100.244 160.496 80.000 170.451 90.626 100.000 70.102 120.943 100.138 140.000 10.000 130.149 80.291 30.534 100.722 80.632 80.331 110.253 150.453 120.487 120.000 40.000 10.479 70.000 120.022 130.000 130.000 50.900 110.128 110.684 30.164 110.413 50.854 110.000 130.000 10.512 170.074 150.003 110.000 40.000 10.000 120.469 160.613 130.132 90.529 80.871 40.227 170.582 80.026 180.787 130.000 60.339 160.000 10.000 80.000 30.626 80.000 50.029 80.000 10.587 100.612 90.411 70.724 100.000 110.000 10.407 70.552 60.513 40.849 110.655 50.408 50.000 120.296 20.000 10.686 160.645 150.145 80.022 90.414 150.633 120.000 10.637 20.224 30.000 70.000 10.650 90.000 20.000 10.622 100.535 130.343 130.483 30.230 140.943 110.289 110.618 80.596 60.140 90.679 90.000 10.022 60.783 120.620 130.906 20.000 10.806 90.137 110.865 60.000 10.378 120.000 160.168 160.680 90.227 14
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
OA-CNN-L_ScanNet2000.333 120.558 60.269 100.124 140.448 150.080 100.272 60.000 30.000 80.000 10.342 90.515 40.524 80.713 140.789 100.158 120.384 130.000 30.806 70.125 80.000 100.496 90.332 70.498 150.227 90.024 70.474 30.000 10.003 20.071 100.487 40.000 120.000 10.110 90.000 30.876 80.013 180.703 40.000 80.076 100.473 130.355 120.906 70.000 10.000 20.476 60.706 10.000 100.672 110.835 140.748 100.015 130.223 80.860 120.000 10.000 110.572 80.000 70.509 70.313 80.662 50.398 140.396 80.411 140.276 20.527 50.711 60.000 70.076 140.946 70.166 60.000 10.022 110.160 70.183 140.493 140.699 100.637 70.403 70.330 130.406 140.526 70.024 20.000 10.392 120.000 120.016 160.000 130.196 40.915 60.112 120.557 110.197 70.352 110.877 40.000 130.000 10.592 130.103 110.000 140.067 10.000 10.089 80.735 80.625 120.130 100.568 70.836 80.271 90.534 100.043 140.799 120.001 50.445 60.000 10.000 80.024 20.661 50.000 50.262 30.000 10.591 90.517 140.373 90.788 70.021 90.000 10.455 50.517 100.320 90.823 130.200 170.001 180.150 60.100 130.000 10.736 100.668 110.103 150.052 70.662 50.720 90.000 10.602 60.112 70.002 60.000 10.637 100.000 20.000 10.621 110.569 60.398 100.412 50.234 130.949 70.363 60.492 150.495 120.251 40.665 100.000 10.001 110.805 80.833 70.794 120.000 10.821 60.314 50.843 120.000 10.560 100.245 80.262 70.713 50.370 11
GSTran0.334 110.533 130.250 130.179 90.487 90.041 170.139 140.003 10.273 60.000 10.273 170.189 170.465 130.704 150.794 90.198 80.831 60.000 30.712 90.055 170.063 60.518 70.306 90.459 170.217 110.028 50.282 150.000 10.000 30.044 130.405 120.558 90.000 10.080 130.000 30.873 100.020 170.684 80.000 80.075 130.496 100.363 100.651 160.000 10.000 20.425 90.648 60.000 100.669 120.914 70.741 110.009 150.200 100.864 100.000 10.000 110.560 100.000 70.357 150.233 130.633 120.408 120.411 40.320 170.242 50.440 100.598 150.047 40.205 90.940 110.139 120.000 10.372 50.138 90.191 110.495 120.618 140.624 100.321 130.595 40.496 70.499 90.000 40.000 10.340 130.014 70.032 70.136 50.000 50.903 90.279 50.601 100.198 60.345 120.849 120.260 70.000 10.573 150.072 170.060 50.000 40.000 10.089 80.838 50.775 40.125 120.381 120.752 140.274 60.517 140.032 160.841 60.000 60.354 150.000 10.047 60.000 30.439 140.787 30.252 40.000 10.512 170.507 170.158 170.717 120.000 110.000 10.337 130.483 140.570 20.853 90.614 80.121 120.070 100.229 80.000 10.732 120.773 20.193 30.000 110.606 80.791 60.000 10.593 90.000 150.010 50.000 10.574 170.000 20.000 10.507 130.554 100.361 120.136 140.608 40.948 80.304 90.593 110.533 90.011 170.634 130.000 10.060 30.821 50.613 140.797 110.000 10.799 120.036 140.782 150.000 10.609 70.423 40.133 180.647 130.213 16
IMFSegNet0.334 100.532 140.251 120.179 80.486 100.041 170.139 140.003 10.283 50.000 10.274 160.191 160.457 150.704 150.795 70.197 90.830 70.000 30.710 100.055 170.064 50.518 70.305 100.458 180.216 130.027 60.284 140.000 10.000 30.044 130.406 110.561 80.000 10.080 130.000 30.873 100.021 160.683 90.000 80.076 100.494 110.363 100.648 170.000 10.000 20.425 90.649 50.000 100.668 130.908 80.740 120.010 140.206 90.862 110.000 10.000 110.560 100.000 70.359 140.237 120.631 130.408 120.411 40.322 160.246 40.439 110.599 140.047 40.213 80.940 110.139 120.000 10.369 60.124 100.188 130.495 120.624 120.626 90.320 150.595 40.495 80.496 110.000 40.000 10.340 130.014 70.032 70.135 60.000 50.903 90.277 60.612 90.196 80.344 130.848 140.260 70.000 10.574 140.073 160.062 40.000 40.000 10.091 70.839 40.776 30.123 130.392 100.756 130.274 60.518 130.029 170.842 50.000 60.357 140.000 10.035 70.000 30.444 130.793 20.245 50.000 10.512 170.512 160.159 160.713 130.000 110.000 10.336 140.484 130.569 30.852 100.615 70.120 130.068 110.228 90.000 10.733 110.773 20.190 40.000 110.608 70.792 50.000 10.597 70.000 150.025 20.000 10.573 180.000 20.000 10.508 120.555 90.363 110.139 130.610 30.947 90.305 80.594 100.527 100.009 180.633 140.000 10.060 30.820 60.604 160.799 100.000 10.799 120.034 150.784 140.000 10.618 60.424 30.134 170.646 140.214 15
PonderV2 ScanNet2000.346 70.552 90.270 90.175 100.497 80.070 130.239 80.000 30.000 80.000 10.232 180.412 90.584 30.842 40.804 50.212 70.540 110.000 30.433 170.106 110.000 100.590 60.290 120.548 60.243 70.000 100.356 120.000 10.000 30.062 110.398 140.441 110.000 10.104 110.000 30.888 60.076 120.682 100.030 40.094 70.491 120.351 130.869 110.000 10.063 10.403 120.700 20.000 100.660 140.881 100.761 40.050 80.186 110.852 140.000 10.007 90.570 90.100 20.565 30.326 70.641 110.431 70.290 150.621 60.259 30.408 120.622 110.125 20.082 130.950 60.179 50.000 10.263 70.424 50.193 100.558 80.880 40.545 140.375 80.727 30.445 130.499 90.000 40.000 10.475 80.002 100.034 60.083 90.000 50.924 30.290 40.636 60.115 150.400 60.874 50.186 110.000 10.611 90.128 30.113 20.000 40.000 10.000 120.584 130.636 100.103 150.385 110.843 70.283 50.603 70.080 80.825 110.000 60.377 110.000 10.000 80.000 30.457 120.000 50.000 90.000 10.574 130.608 100.481 40.792 50.394 60.000 10.357 110.503 120.261 110.817 140.504 130.304 80.472 50.115 120.000 10.750 80.677 100.202 20.000 110.509 100.729 70.000 10.519 130.000 150.000 70.000 10.620 130.000 20.000 10.660 70.560 80.486 70.384 60.346 110.952 60.247 150.667 50.436 130.269 30.691 70.000 10.010 70.787 110.889 30.880 50.000 10.810 80.336 40.860 90.000 10.606 80.009 120.248 100.681 80.392 9
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.
BFANet ScanNet200permissive0.360 60.553 80.293 60.193 60.483 110.096 70.266 70.000 30.000 80.000 10.298 140.255 130.661 10.810 60.810 30.194 100.785 80.000 30.000 180.161 70.000 100.494 100.382 30.574 40.258 50.000 100.372 100.000 10.000 30.043 150.436 90.000 120.000 10.239 30.000 30.901 30.105 10.689 50.025 50.128 40.614 30.436 20.493 180.000 10.000 20.526 40.546 140.109 50.651 150.953 60.753 70.101 50.143 140.897 60.000 10.431 10.469 160.000 70.522 60.337 60.661 70.459 40.409 60.666 50.102 140.508 70.757 50.000 70.060 150.970 30.497 10.000 10.376 40.511 30.262 50.688 30.921 20.617 110.321 130.590 60.491 90.556 50.000 40.000 10.481 60.093 10.043 30.284 20.000 50.875 150.135 90.669 40.124 140.394 70.849 120.298 50.000 10.476 180.088 130.042 70.000 40.000 10.254 50.653 110.741 60.215 10.573 60.852 60.266 110.654 30.056 130.835 70.000 60.492 20.000 10.000 80.000 30.612 100.000 50.000 90.000 10.616 70.469 180.460 50.698 140.516 30.000 10.378 90.563 50.476 50.863 60.574 100.330 70.000 120.282 30.000 10.760 50.710 60.233 10.000 110.641 60.814 30.000 10.585 100.053 120.000 70.000 10.629 110.000 20.000 10.678 40.528 140.534 60.129 150.596 50.973 50.264 130.772 20.526 110.139 100.707 50.000 10.000 120.764 150.591 170.848 70.000 10.827 50.338 30.806 130.000 10.568 90.151 110.358 20.659 110.510 4
Weiguang Zhao, Rui Zhang, Qiufeng Wang, Guangliang Cheng, Kaizhu Huang: BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis. CVPR 2025
Minkowski 34Dpermissive0.253 170.463 170.154 180.102 170.381 170.084 90.134 160.000 30.000 80.000 10.386 70.141 180.279 180.737 130.703 170.014 180.164 160.000 30.663 110.092 150.000 100.224 160.291 110.531 90.056 180.000 100.242 170.000 10.000 30.013 160.331 170.000 120.000 10.035 180.001 20.858 150.059 150.650 170.000 80.056 150.353 160.299 160.670 140.000 10.000 20.284 170.484 160.071 80.594 160.720 170.710 160.027 110.068 180.813 150.000 10.005 100.492 150.164 10.274 170.111 170.571 170.307 180.293 140.307 180.150 90.163 180.531 170.002 60.545 50.932 160.093 180.000 10.000 130.002 140.159 160.368 180.581 160.440 180.228 180.406 100.282 180.294 170.000 40.000 10.189 170.060 20.036 50.000 130.000 50.897 120.000 180.525 150.025 180.205 180.771 180.000 130.000 10.593 120.108 90.044 60.000 40.000 10.000 120.282 180.589 150.094 170.169 170.466 180.227 170.419 180.125 50.757 150.002 40.334 170.000 10.000 80.000 30.357 160.000 50.000 90.000 10.582 110.513 150.337 110.612 180.000 110.000 10.250 170.352 180.136 180.724 170.655 50.280 90.000 120.046 170.000 10.606 180.559 160.159 70.102 30.445 110.655 100.000 10.310 180.117 60.000 70.000 10.581 160.026 10.000 10.265 180.483 170.084 180.097 180.044 160.865 180.142 180.588 120.351 160.272 20.596 180.000 10.003 100.622 170.720 110.096 180.000 10.771 170.016 160.772 160.000 10.302 150.194 100.214 130.621 170.197 17
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
ODIN - Sem200permissive0.368 50.562 50.297 50.207 50.380 180.196 10.828 30.000 30.321 30.000 10.400 50.775 10.460 140.501 180.769 130.065 160.870 40.000 30.913 10.213 40.000 100.000 180.389 20.554 50.312 30.000 100.591 10.000 10.000 30.491 10.487 40.894 20.000 10.378 20.303 10.796 180.088 60.669 140.081 20.216 10.256 180.334 140.898 80.000 10.000 20.370 150.599 110.000 100.581 170.988 20.749 90.090 60.242 60.921 40.000 10.202 50.609 30.000 70.655 10.214 140.654 100.346 160.408 70.485 90.169 80.631 20.704 70.000 70.814 10.940 110.127 170.000 10.000 130.462 40.227 70.641 50.885 30.657 60.434 30.000 180.550 20.393 160.000 40.000 10.590 40.000 120.048 20.077 100.000 50.784 170.131 100.557 110.316 20.359 90.833 150.373 30.000 10.661 40.108 90.001 120.000 40.000 10.301 40.612 120.565 160.129 110.482 90.468 170.274 60.561 90.376 10.912 20.181 10.440 70.000 10.166 40.000 30.641 60.000 50.426 20.000 10.642 60.626 80.259 120.787 80.429 50.000 10.589 10.523 90.246 120.857 70.000 180.228 100.000 120.265 40.000 10.752 70.832 10.090 170.157 10.791 20.578 170.000 10.373 160.539 10.000 70.000 10.685 50.000 20.000 10.632 90.575 40.663 10.152 120.358 100.926 140.397 40.454 160.610 50.119 160.685 80.000 10.000 120.803 90.740 100.441 150.000 10.800 110.000 180.871 40.000 10.220 180.487 20.862 10.682 70.054 18
Ayush Jain, Pushkal Katara, Nikolaos Gkanatsios, Adam W. Harley, Gabriel Sarch, Kriti Aggarwal, Vishrav Chaudhary, Katerina Fragkiadaki: ODIN: A Single Model for 2D and 3D Segmentation. CVPR 2024
CSC-Pretrainpermissive0.249 180.455 180.171 170.079 180.418 160.059 150.186 110.000 30.000 80.000 10.335 110.250 140.316 170.766 80.697 180.142 140.170 150.003 20.553 150.112 100.097 10.201 170.186 150.476 160.081 170.000 100.216 180.000 10.000 30.001 180.314 180.000 120.000 10.055 160.000 30.832 170.094 30.659 160.002 60.076 100.310 170.293 180.664 150.000 10.000 20.175 180.634 70.130 20.552 180.686 180.700 180.076 70.110 160.770 180.000 10.000 110.430 180.000 70.319 160.166 160.542 180.327 170.205 170.332 150.052 160.375 140.444 180.000 70.012 180.930 180.203 30.000 10.000 130.046 120.175 150.413 170.592 150.471 170.299 160.152 170.340 170.247 180.000 40.000 10.225 160.058 30.037 40.000 130.207 30.862 160.014 140.548 140.033 170.233 170.816 170.000 130.000 10.542 160.123 50.121 10.019 20.000 10.000 120.463 170.454 180.045 180.128 180.557 160.235 150.441 170.063 110.484 180.000 60.308 180.000 10.000 80.000 30.318 180.000 50.000 90.000 10.545 150.543 130.164 150.734 90.000 110.000 10.215 180.371 170.198 150.743 150.205 160.062 160.000 120.079 150.000 10.683 170.547 170.142 90.000 110.441 120.579 160.000 10.464 150.098 90.041 10.000 10.590 150.000 20.000 10.373 140.494 150.174 160.105 170.001 180.895 170.222 170.537 130.307 170.180 50.625 150.000 10.000 120.591 180.609 150.398 160.000 10.766 180.014 170.638 180.000 10.377 130.004 140.206 140.609 180.465 5
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021