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 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 bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
DITR0.409 20.616 10.351 10.215 30.651 10.238 10.400 20.000 10.340 10.000 10.534 20.476 40.585 20.687 140.853 10.143 110.854 20.000 30.865 30.167 50.000 80.175 150.573 10.617 20.372 10.362 10.591 10.000 10.000 30.330 10.494 20.247 80.000 10.385 10.000 20.878 60.037 140.791 10.053 20.118 30.479 100.429 40.940 30.000 10.000 20.461 80.562 90.093 50.628 130.991 10.762 30.135 30.270 30.917 30.000 10.140 40.597 20.000 70.361 120.375 10.730 20.431 50.459 30.410 130.008 140.656 10.814 10.036 40.554 40.947 60.139 110.000 10.263 30.896 10.191 90.615 40.839 30.757 10.399 60.877 10.504 50.524 60.000 40.000 10.587 30.000 80.022 90.077 80.921 10.928 20.132 80.670 40.759 10.652 10.862 70.091 90.000 10.662 30.072 150.000 110.000 40.000 10.496 10.852 20.752 20.152 30.743 10.953 10.301 30.625 30.053 120.913 10.399 10.452 50.000 10.000 60.000 30.742 20.000 20.000 60.000 10.694 20.643 40.444 60.784 70.000 90.000 10.571 10.614 30.491 20.938 10.559 90.357 50.107 70.404 10.000 10.796 20.688 40.148 60.186 10.629 50.827 20.000 10.558 100.198 40.000 60.000 10.723 20.000 20.000 10.833 10.619 10.609 20.478 40.617 10.959 40.370 30.597 90.737 20.191 50.752 20.000 10.118 10.853 10.925 20.670 120.000 10.831 30.000 150.873 30.000 10.699 10.005 100.360 10.723 30.235 13
ALS-MinkowskiNetcopyleft0.414 10.610 20.322 30.271 10.542 20.153 20.159 110.000 10.000 40.000 10.404 40.503 30.532 60.672 150.804 50.285 20.888 10.000 30.900 10.226 10.087 20.598 30.342 40.671 10.217 90.087 30.449 30.000 10.000 30.253 20.477 51.000 10.000 10.118 40.000 20.905 10.071 110.710 20.076 10.047 140.665 10.376 80.981 10.000 10.000 20.466 70.632 60.113 30.769 10.956 30.795 10.031 100.314 10.936 10.000 10.390 20.601 10.000 70.458 70.366 20.719 30.440 40.564 10.699 30.314 20.464 70.784 20.200 10.283 60.973 10.142 90.000 10.250 50.285 60.220 50.718 10.752 50.723 20.460 10.248 140.475 70.463 120.000 40.000 10.446 70.021 40.025 70.285 10.000 50.972 10.149 60.769 10.230 20.535 20.879 20.252 40.000 10.693 10.129 20.000 110.000 40.000 10.447 20.958 10.662 80.159 20.598 30.780 120.344 20.646 20.106 40.893 20.135 20.455 40.000 10.194 30.259 10.726 30.475 10.000 60.000 10.741 10.865 10.571 20.817 30.445 30.000 10.506 20.630 20.230 100.916 20.728 10.635 11.000 10.252 50.000 10.804 10.697 30.137 90.043 60.717 20.807 40.000 10.510 120.245 10.000 60.000 10.709 30.000 20.000 10.703 20.572 30.646 10.223 110.531 30.984 10.397 20.813 10.798 10.135 130.800 10.000 10.097 20.832 30.752 80.842 80.000 10.852 10.149 90.846 90.000 10.666 50.359 20.252 80.777 10.690 2
OA-CNN-L_ScanNet2000.333 90.558 40.269 90.124 110.448 130.080 80.272 50.000 10.000 40.000 10.342 70.515 20.524 70.713 130.789 70.158 100.384 100.000 30.806 60.125 70.000 80.496 50.332 60.498 140.227 70.024 40.474 20.000 10.003 20.071 80.487 30.000 90.000 10.110 70.000 20.876 70.013 150.703 30.000 70.076 90.473 110.355 100.906 60.000 10.000 20.476 50.706 10.000 100.672 100.835 110.748 90.015 140.223 70.860 90.000 10.000 100.572 60.000 70.509 60.313 70.662 50.398 110.396 50.411 120.276 30.527 30.711 50.000 70.076 110.946 70.166 60.000 10.022 90.160 70.183 110.493 110.699 80.637 60.403 50.330 110.406 110.526 50.024 30.000 10.392 100.000 80.016 140.000 100.196 30.915 60.112 100.557 80.197 40.352 90.877 30.000 100.000 10.592 120.103 100.000 110.067 10.000 10.089 50.735 50.625 100.130 90.568 60.836 80.271 60.534 90.043 130.799 80.001 50.445 60.000 10.000 60.024 20.661 40.000 20.262 20.000 10.591 70.517 130.373 80.788 60.021 80.000 10.455 30.517 80.320 70.823 90.200 150.001 140.150 50.100 90.000 10.736 90.668 70.103 120.052 50.662 30.720 70.000 10.602 60.112 60.002 50.000 10.637 80.000 20.000 10.621 90.569 40.398 90.412 60.234 90.949 60.363 50.492 130.495 90.251 40.665 90.000 10.001 100.805 60.833 60.794 90.000 10.821 50.314 50.843 100.000 10.560 80.245 30.262 60.713 40.370 11
GSTran0.339 70.536 100.273 60.169 70.491 80.071 110.365 30.000 10.000 40.000 10.178 150.246 130.458 120.754 80.788 80.316 10.834 40.000 30.872 20.202 30.079 30.318 120.286 100.538 70.156 120.004 70.310 110.000 10.000 30.009 130.397 110.297 70.000 10.093 100.000 20.876 70.060 120.690 40.000 70.086 70.517 70.358 90.667 130.000 10.000 20.473 60.670 30.000 100.731 30.896 60.765 20.061 80.256 50.889 60.000 10.000 100.480 120.000 70.412 110.279 80.690 40.366 120.373 60.466 90.357 10.514 50.648 80.024 50.615 20.949 50.183 40.000 10.162 70.564 30.196 70.535 80.413 130.638 50.410 40.682 40.445 90.470 110.289 20.000 10.358 110.000 80.022 90.161 40.008 40.877 120.495 10.461 140.161 90.348 100.853 110.199 60.000 10.643 40.109 70.014 80.000 40.000 10.000 80.681 60.705 60.079 140.441 80.872 30.282 50.593 70.096 50.786 100.021 30.495 10.000 10.118 40.000 30.487 100.000 20.002 50.000 10.589 80.563 110.144 140.682 120.109 70.000 10.235 140.455 110.368 50.659 150.609 60.000 150.060 90.033 140.000 10.746 70.648 110.084 140.000 90.803 10.832 10.000 10.614 40.000 130.497 10.000 10.597 120.000 20.000 10.621 90.506 110.459 60.252 100.228 110.913 120.369 40.665 50.598 40.139 100.666 80.000 10.097 20.841 20.698 100.857 60.000 10.811 70.129 110.784 120.000 10.386 100.012 70.317 30.696 50.425 7
BFANet ScanNet200permissive0.360 40.553 60.293 40.193 40.483 90.096 50.266 60.000 10.000 40.000 10.298 120.255 110.661 10.810 50.810 30.194 80.785 50.000 30.000 150.161 60.000 80.494 60.382 20.574 40.258 40.000 80.372 80.000 10.000 30.043 110.436 70.000 90.000 10.239 20.000 20.901 30.105 10.689 50.025 40.128 20.614 20.436 10.493 150.000 10.000 20.526 40.546 110.109 40.651 120.953 40.753 70.101 60.143 110.897 40.000 10.431 10.469 130.000 70.522 50.337 50.661 70.459 20.409 40.666 40.102 110.508 60.757 40.000 70.060 120.970 30.497 10.000 10.376 20.511 40.262 40.688 20.921 10.617 80.321 120.590 50.491 60.556 30.000 40.000 10.481 40.093 10.043 20.284 20.000 50.875 130.135 70.669 50.124 110.394 60.849 120.298 20.000 10.476 150.088 120.042 50.000 40.000 10.254 30.653 90.741 40.215 10.573 50.852 60.266 80.654 10.056 110.835 30.000 60.492 20.000 10.000 60.000 30.612 80.000 20.000 60.000 10.616 50.469 150.460 40.698 110.516 20.000 10.378 70.563 40.476 30.863 50.574 80.330 60.000 100.282 40.000 10.760 40.710 20.233 10.000 90.641 40.814 30.000 10.585 80.053 100.000 60.000 10.629 90.000 20.000 10.678 30.528 100.534 40.129 120.596 20.973 30.264 100.772 20.526 80.139 100.707 40.000 10.000 110.764 120.591 140.848 70.000 10.827 40.338 30.806 110.000 10.568 70.151 60.358 20.659 100.510 4
L3DETR-ScanNet_2000.336 80.533 110.279 50.155 80.508 50.073 100.101 150.000 10.058 30.000 10.294 130.233 140.548 40.927 10.788 80.264 30.463 90.000 30.638 100.098 130.014 60.411 90.226 110.525 110.225 80.010 50.397 50.000 10.000 30.192 50.380 120.598 40.000 10.117 50.000 20.883 50.082 70.689 50.000 70.032 150.549 50.417 50.910 50.000 10.000 20.448 90.613 80.000 100.697 70.960 20.759 50.158 20.293 20.883 70.000 10.312 30.583 30.079 40.422 100.068 150.660 80.418 70.298 100.430 110.114 90.526 40.776 30.051 30.679 10.946 70.152 70.000 10.183 60.000 130.211 60.511 100.409 140.565 100.355 80.448 70.512 40.557 20.000 40.000 10.420 80.000 80.007 150.104 50.000 50.125 150.330 30.514 120.146 100.321 110.860 80.174 80.000 10.629 60.075 130.000 110.000 40.000 10.002 70.671 70.712 50.141 60.339 100.856 50.261 100.529 100.067 90.835 30.000 60.369 120.000 10.259 20.000 30.629 50.000 20.487 10.000 10.579 110.646 30.107 150.720 100.122 60.000 10.333 110.505 90.303 80.908 30.503 130.565 20.074 80.324 20.000 10.740 80.661 80.109 110.000 90.427 110.563 150.000 10.579 90.108 70.000 60.000 10.664 50.000 20.000 10.641 70.539 80.416 70.515 20.256 80.940 100.312 60.209 150.620 30.138 120.636 110.000 10.000 110.775 110.861 50.765 100.000 10.801 100.119 120.860 70.000 10.687 20.001 120.192 140.679 90.699 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12
PTv3 ScanNet2000.393 30.592 30.330 20.216 20.520 30.109 40.108 140.000 10.337 20.000 10.310 110.394 80.494 110.753 90.848 20.256 40.717 60.000 30.842 40.192 40.065 40.449 70.346 30.546 60.190 100.000 80.384 60.000 10.000 30.218 30.505 10.791 20.000 10.136 30.000 20.903 20.073 100.687 70.000 70.168 10.551 40.387 70.941 20.000 10.000 20.397 110.654 40.000 100.714 50.759 130.752 80.118 50.264 40.926 20.000 10.048 50.575 40.000 70.597 10.366 20.755 10.469 10.474 20.798 10.140 80.617 20.692 60.000 70.592 30.971 20.188 30.000 10.133 80.593 20.349 10.650 30.717 70.699 30.455 20.790 20.523 30.636 10.301 10.000 10.622 20.000 80.017 130.259 30.000 50.921 40.337 20.733 20.210 30.514 30.860 80.407 10.000 10.688 20.109 70.000 110.000 40.000 10.151 40.671 70.782 10.115 100.641 20.903 20.349 10.616 40.088 60.832 50.000 60.480 30.000 10.428 10.000 30.497 90.000 20.000 60.000 10.662 40.690 20.612 10.828 10.575 10.000 10.404 60.644 10.325 60.887 40.728 10.009 130.134 60.026 150.000 10.761 30.731 10.172 40.077 30.528 60.727 60.000 10.603 50.220 30.022 30.000 10.740 10.000 20.000 10.661 40.586 20.566 30.436 50.531 30.978 20.457 10.708 30.583 60.141 80.748 30.000 10.026 40.822 40.871 40.879 50.000 10.851 20.405 20.914 10.000 10.682 30.000 130.281 40.738 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)
PonderV2 ScanNet2000.346 50.552 70.270 80.175 60.497 70.070 120.239 70.000 10.000 40.000 10.232 140.412 70.584 30.842 30.804 50.212 70.540 80.000 30.433 140.106 100.000 80.590 40.290 90.548 50.243 60.000 80.356 100.000 10.000 30.062 90.398 100.441 60.000 10.104 90.000 20.888 40.076 90.682 80.030 30.094 60.491 90.351 110.869 90.000 10.063 10.403 100.700 20.000 100.660 110.881 70.761 40.050 90.186 80.852 110.000 10.007 80.570 70.100 20.565 20.326 60.641 100.431 50.290 120.621 50.259 40.408 90.622 100.125 20.082 100.950 40.179 50.000 10.263 30.424 50.193 80.558 60.880 20.545 110.375 70.727 30.445 90.499 80.000 40.000 10.475 60.002 60.034 50.083 70.000 50.924 30.290 40.636 60.115 120.400 50.874 40.186 70.000 10.611 80.128 30.113 20.000 40.000 10.000 80.584 100.636 90.103 110.385 90.843 70.283 40.603 60.080 70.825 70.000 60.377 100.000 10.000 60.000 30.457 110.000 20.000 60.000 10.574 120.608 80.481 30.792 40.394 40.000 10.357 90.503 100.261 90.817 100.504 120.304 70.472 40.115 80.000 10.750 60.677 60.202 20.000 90.509 70.729 50.000 10.519 110.000 130.000 60.000 10.620 110.000 20.000 10.660 60.560 60.486 50.384 70.346 70.952 50.247 120.667 40.436 100.269 30.691 60.000 10.010 60.787 80.889 30.880 40.000 10.810 80.336 40.860 70.000 10.606 60.009 80.248 90.681 70.392 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.
PPT-SpUNet-F.T.0.332 100.556 50.270 70.123 120.519 40.091 60.349 40.000 10.000 40.000 10.339 80.383 90.498 100.833 40.807 40.241 50.584 70.000 30.755 70.124 80.000 80.608 20.330 70.530 100.314 20.000 80.374 70.000 10.000 30.197 40.459 60.000 90.000 10.117 50.000 20.876 70.095 20.682 80.000 70.086 70.518 60.433 20.930 40.000 10.000 20.563 30.542 120.077 70.715 40.858 90.756 60.008 150.171 100.874 80.000 10.039 60.550 80.000 70.545 40.256 90.657 90.453 30.351 80.449 100.213 50.392 100.611 110.000 70.037 130.946 70.138 120.000 10.000 110.063 90.308 20.537 70.796 40.673 40.323 110.392 90.400 120.509 70.000 40.000 10.649 10.000 80.023 80.000 100.000 50.914 70.002 140.506 130.163 80.359 80.872 50.000 100.000 10.623 70.112 50.001 100.000 40.000 10.021 60.753 30.565 140.150 40.579 40.806 100.267 70.616 40.042 140.783 110.000 60.374 110.000 10.000 60.000 30.620 70.000 20.000 60.000 10.572 130.634 50.350 90.792 40.000 90.000 10.376 80.535 60.378 40.855 60.672 30.074 100.000 100.185 70.000 10.727 100.660 90.076 150.000 90.432 100.646 90.000 10.594 70.006 120.000 60.000 10.658 60.000 20.000 10.661 40.549 70.300 120.291 90.045 120.942 90.304 70.600 80.572 70.135 130.695 50.000 10.008 80.793 70.942 10.899 20.000 10.816 60.181 70.897 20.000 10.679 40.223 40.264 50.691 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
LGroundpermissive0.272 130.485 130.184 130.106 130.476 100.077 90.218 80.000 10.000 40.000 10.547 10.295 100.540 50.746 100.745 130.058 140.112 140.005 10.658 90.077 150.000 80.322 110.178 140.512 120.190 100.199 20.277 130.000 10.000 30.173 60.399 90.000 90.000 10.039 140.000 20.858 130.085 60.676 100.002 50.103 50.498 80.323 120.703 110.000 10.000 20.296 130.549 100.216 10.702 60.768 120.718 120.028 110.092 140.786 140.000 10.000 100.453 140.022 50.251 150.252 100.572 130.348 130.321 90.514 60.063 120.279 140.552 130.000 70.019 140.932 130.132 140.000 10.000 110.000 130.156 150.457 130.623 100.518 120.265 140.358 100.381 130.395 130.000 40.000 10.127 150.012 50.051 10.000 100.000 50.886 110.014 120.437 150.179 50.244 130.826 130.000 100.000 10.599 100.136 10.085 30.000 40.000 10.000 80.565 110.612 120.143 50.207 130.566 130.232 130.446 130.127 20.708 130.000 60.384 90.000 10.000 60.000 30.402 120.000 20.059 30.000 10.525 150.566 100.229 110.659 130.000 90.000 10.265 120.446 120.147 140.720 140.597 70.066 110.000 100.187 60.000 10.726 110.467 150.134 100.000 90.413 130.629 110.000 10.363 140.055 90.022 30.000 10.626 100.000 20.000 10.323 130.479 150.154 140.117 130.028 140.901 130.243 130.415 140.295 150.143 70.610 140.000 10.000 110.777 100.397 150.324 140.000 10.778 130.179 80.702 140.000 10.274 150.404 10.233 100.622 130.398 8
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
OctFormer ScanNet200permissive0.326 110.539 90.265 100.131 100.499 60.110 30.522 10.000 10.000 40.000 10.318 100.427 60.455 130.743 110.765 110.175 90.842 30.000 30.828 50.204 20.033 50.429 80.335 50.601 30.312 30.000 80.357 90.000 10.000 30.047 100.423 80.000 90.000 10.105 80.000 20.873 100.079 80.670 110.000 70.117 40.471 120.432 30.829 100.000 10.000 20.584 20.417 150.089 60.684 90.837 100.705 140.021 130.178 90.892 50.000 10.028 70.505 100.000 70.457 80.200 120.662 50.412 90.244 130.496 70.000 150.451 80.626 90.000 70.102 90.943 100.138 120.000 10.000 110.149 80.291 30.534 90.722 60.632 70.331 100.253 130.453 80.487 90.000 40.000 10.479 50.000 80.022 90.000 100.000 50.900 80.128 90.684 30.164 70.413 40.854 100.000 100.000 10.512 140.074 140.003 90.000 40.000 10.000 80.469 130.613 110.132 80.529 70.871 40.227 140.582 80.026 150.787 90.000 60.339 130.000 10.000 60.000 30.626 60.000 20.029 40.000 10.587 90.612 70.411 70.724 90.000 90.000 10.407 50.552 50.513 10.849 70.655 40.408 30.000 100.296 30.000 10.686 130.645 120.145 70.022 70.414 120.633 100.000 10.637 10.224 20.000 60.000 10.650 70.000 20.000 10.622 80.535 90.343 100.483 30.230 100.943 80.289 80.618 70.596 50.140 90.679 70.000 10.022 50.783 90.620 120.906 10.000 10.806 90.137 100.865 40.000 10.378 110.000 130.168 150.680 80.227 14
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
CeCo0.340 60.551 80.247 110.181 50.475 110.057 150.142 120.000 10.000 40.000 10.387 50.463 50.499 90.924 20.774 100.213 60.257 110.000 30.546 130.100 110.006 70.615 10.177 150.534 80.246 50.000 80.400 40.000 10.338 10.006 140.484 40.609 30.000 10.083 110.000 20.873 100.089 50.661 120.000 70.048 130.560 30.408 60.892 70.000 10.000 20.586 10.616 70.000 100.692 80.900 50.721 100.162 10.228 60.860 90.000 10.000 100.575 40.083 30.550 30.347 40.624 110.410 100.360 70.740 20.109 100.321 130.660 70.000 70.121 70.939 110.143 80.000 10.400 10.003 110.190 100.564 50.652 90.615 90.421 30.304 120.579 10.547 40.000 40.000 10.296 120.000 80.030 60.096 60.000 50.916 50.037 110.551 90.171 60.376 70.865 60.286 30.000 10.633 50.102 110.027 60.011 30.000 10.000 80.474 120.742 30.133 70.311 110.824 90.242 110.503 120.068 80.828 60.000 60.429 70.000 10.063 50.000 30.781 10.000 20.000 60.000 10.665 30.633 60.450 50.818 20.000 90.000 10.429 40.532 70.226 110.825 80.510 110.377 40.709 20.079 110.000 10.753 50.683 50.102 130.063 40.401 140.620 120.000 10.619 20.000 130.000 60.000 10.595 130.000 20.000 10.345 120.564 50.411 80.603 10.384 60.945 70.266 90.643 60.367 120.304 10.663 100.000 10.010 60.726 130.767 70.898 30.000 10.784 110.435 10.861 60.000 10.447 90.000 130.257 70.656 110.377 10
Zhisheng Zhong, Jiequan Cui, Yibo Yang, Xiaoyang Wu, Xiaojuan Qi, Xiangyu Zhang, Jiaya Jia: Understanding Imbalanced Semantic Segmentation Through Neural Collapse. CVPR 2023
CSC-Pretrainpermissive0.249 150.455 150.171 140.079 150.418 140.059 140.186 100.000 10.000 40.000 10.335 90.250 120.316 140.766 70.697 150.142 120.170 120.003 20.553 120.112 90.097 10.201 140.186 120.476 150.081 140.000 80.216 150.000 10.000 30.001 150.314 150.000 90.000 10.055 130.000 20.832 150.094 30.659 130.002 50.076 90.310 150.293 150.664 140.000 10.000 20.175 150.634 50.130 20.552 150.686 150.700 150.076 70.110 130.770 150.000 10.000 100.430 150.000 70.319 130.166 130.542 150.327 140.205 140.332 140.052 130.375 110.444 150.000 70.012 150.930 150.203 20.000 10.000 110.046 100.175 120.413 140.592 110.471 140.299 130.152 150.340 140.247 150.000 40.000 10.225 130.058 30.037 30.000 100.207 20.862 140.014 120.548 100.033 140.233 140.816 140.000 100.000 10.542 130.123 40.121 10.019 20.000 10.000 80.463 140.454 150.045 150.128 150.557 140.235 120.441 140.063 100.484 150.000 60.308 150.000 10.000 60.000 30.318 150.000 20.000 60.000 10.545 140.543 120.164 130.734 80.000 90.000 10.215 150.371 140.198 120.743 110.205 140.062 120.000 100.079 110.000 10.683 140.547 140.142 80.000 90.441 90.579 140.000 10.464 130.098 80.041 20.000 10.590 140.000 20.000 10.373 110.494 120.174 130.105 140.001 150.895 140.222 140.537 110.307 140.180 60.625 120.000 10.000 110.591 150.609 130.398 130.000 10.766 150.014 140.638 150.000 10.377 120.004 110.206 130.609 150.465 5
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
Minkowski 34Dpermissive0.253 140.463 140.154 150.102 140.381 150.084 70.134 130.000 10.000 40.000 10.386 60.141 150.279 150.737 120.703 140.014 150.164 130.000 30.663 80.092 140.000 80.224 130.291 80.531 90.056 150.000 80.242 140.000 10.000 30.013 120.331 140.000 90.000 10.035 150.001 10.858 130.059 130.650 140.000 70.056 120.353 140.299 130.670 120.000 10.000 20.284 140.484 130.071 80.594 140.720 140.710 130.027 120.068 150.813 120.000 10.005 90.492 110.164 10.274 140.111 140.571 140.307 150.293 110.307 150.150 70.163 150.531 140.002 60.545 50.932 130.093 150.000 10.000 110.002 120.159 130.368 150.581 120.440 150.228 150.406 80.282 150.294 140.000 40.000 10.189 140.060 20.036 40.000 100.000 50.897 90.000 150.525 110.025 150.205 150.771 150.000 100.000 10.593 110.108 90.044 40.000 40.000 10.000 80.282 150.589 130.094 130.169 140.466 150.227 140.419 150.125 30.757 120.002 40.334 140.000 10.000 60.000 30.357 130.000 20.000 60.000 10.582 100.513 140.337 100.612 150.000 90.000 10.250 130.352 150.136 150.724 130.655 40.280 80.000 100.046 130.000 10.606 150.559 130.159 50.102 20.445 80.655 80.000 10.310 150.117 50.000 60.000 10.581 150.026 10.000 10.265 150.483 140.084 150.097 150.044 130.865 150.142 150.588 100.351 130.272 20.596 150.000 10.003 90.622 140.720 90.096 150.000 10.771 140.016 130.772 130.000 10.302 130.194 50.214 120.621 140.197 15
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
AWCS0.305 120.508 120.225 120.142 90.463 120.063 130.195 90.000 10.000 40.000 10.467 30.551 10.504 80.773 60.764 120.142 120.029 150.000 30.626 110.100 110.000 80.360 100.179 130.507 130.137 130.006 60.300 120.000 10.000 30.172 70.364 130.512 50.000 10.056 120.000 20.865 120.093 40.634 150.000 70.071 110.396 130.296 140.876 80.000 10.000 20.373 120.436 140.063 90.749 20.877 80.721 100.131 40.124 120.804 130.000 10.000 100.515 90.010 60.452 90.252 100.578 120.417 80.179 150.484 80.171 60.337 120.606 120.000 70.115 80.937 120.142 90.000 10.008 100.000 130.157 140.484 120.402 150.501 130.339 90.553 60.529 20.478 100.000 40.000 10.404 90.001 70.022 90.077 80.000 50.894 100.219 50.628 70.093 130.305 120.886 10.233 50.000 10.603 90.112 50.023 70.000 40.000 10.000 80.741 40.664 70.097 120.253 120.782 110.264 90.523 110.154 10.707 140.000 60.411 80.000 10.000 60.000 30.332 140.000 20.000 60.000 10.602 60.595 90.185 120.656 140.159 50.000 10.355 100.424 130.154 130.729 120.516 100.220 90.620 30.084 100.000 10.707 120.651 100.173 30.014 80.381 150.582 130.000 10.619 20.049 110.000 60.000 10.702 40.000 20.000 10.302 140.489 130.317 110.334 80.392 50.922 110.254 110.533 120.394 110.129 150.613 130.000 10.000 110.820 50.649 110.749 110.000 10.782 120.282 60.863 50.000 10.288 140.006 90.220 110.633 120.542 3