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 iouwallchairfloortabledoorcouchcabinetshelfdeskoffice chairbedpillowsinkpicturewindowtoiletbookshelfmonitorcurtainbookarmchaircoffee tableboxrefrigeratorlampkitchen cabinettowelclothestvnightstandcounterdresserstoolcushionplantceilingbathtubend tabledining tablekeyboardbagbackpacktoilet paperprintertv standwhiteboardblanketshower curtaintrash canclosetstairsmicrowavestoveshoecomputer towerbottlebinottomanbenchboardwashing machinemirrorcopierbasketsofa chairfile cabinetfanlaptopshowerpaperpersonpaper towel dispenserovenblindsrackplateblackboardpianosuitcaserailradiatorrecycling bincontainerwardrobesoap dispensertelephonebucketclockstandlightlaundry basketpipeclothes dryerguitartoilet paper holderseatspeakercolumnbicycleladderbathroom stallshower wallcupjacketstorage bincoffee makerdishwasherpaper towel rollmachinematwindowsillbartoasterbulletin boardironing boardfireplacesoap dishkitchen counterdoorframetoilet paper dispensermini fridgefire extinguisherballhatshower curtain rodwater coolerpaper cuttertrayshower doorpillarledgetoaster ovenmousetoilet seat cover dispenserfurniturecartstorage containerscaletissue boxlight switchcratepower outletdecorationsignprojectorcloset doorvacuum cleanercandleplungerstuffed animalheadphonesdish rackbroomguitar caserange hooddustpanhair dryerwater bottlehandicap barpurseventshower floorwater pitchermailboxbowlpaper bagalarm clockmusic standprojector screendividerlaundry detergentbathroom counterobjectbathroom vanitycloset walllaundry hamperbathroom stall doorceiling lighttrash bindumbbellstair railtubebathroom cabinetcd casecloset rodcoffee kettlestructureshower headkeyboard pianocase of water bottlescoat rackstorage organizerfolded chairfire alarmpower stripcalendarposterpotted plantluggagemattress
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 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 by
BFANet ScanNet200permissive0.360 30.553 50.293 30.193 30.827 30.689 30.970 30.528 90.661 50.753 50.436 60.378 60.469 110.042 50.810 20.654 10.760 30.266 60.659 80.973 30.574 30.849 100.897 30.382 10.546 90.372 70.698 100.491 50.617 60.526 60.436 10.764 100.476 130.101 50.409 30.585 70.000 10.835 20.901 30.810 50.102 100.000 70.688 20.096 40.483 70.264 80.612 70.591 120.358 10.161 40.863 40.707 30.128 20.814 10.669 40.629 80.563 30.651 110.258 30.000 30.194 70.494 60.806 100.394 50.953 30.000 30.233 10.757 30.508 40.556 30.476 20.000 10.573 40.741 30.000 40.000 70.000 10.000 40.000 130.852 40.678 20.616 40.460 40.338 30.710 20.534 30.000 30.025 30.000 10.043 20.000 30.056 100.493 130.000 10.000 80.109 40.785 30.590 30.298 110.282 30.143 90.262 40.053 90.526 40.337 40.215 10.000 30.135 60.510 40.000 10.596 10.043 100.511 20.321 100.459 20.772 20.124 90.060 100.266 40.000 10.574 70.568 60.653 70.000 10.093 10.298 20.239 10.000 50.516 20.129 100.284 20.000 50.431 10.000 10.000 20.848 60.000 10.492 10.000 10.376 20.522 50.000 10.469 130.000 10.000 10.330 50.151 60.875 110.000 60.254 20.000 10.000 50.000 10.088 110.661 10.481 30.255 100.105 10.139 90.666 40.641 30.000 90.000 10.614 20.000 20.000 20.000 80.921 10.000 20.000 10.000 10.497 10.000 50.000 30.000 70.000 1
ALS-MinkowskiNetcopyleft0.414 10.610 10.322 20.271 10.852 10.710 10.973 10.572 20.719 20.795 10.477 40.506 10.601 10.000 100.804 40.646 20.804 10.344 20.777 10.984 10.671 10.879 20.936 10.342 30.632 50.449 20.817 30.475 60.723 10.798 10.376 70.832 10.693 10.031 80.564 10.510 100.000 10.893 10.905 10.672 130.314 10.000 70.718 10.153 10.542 10.397 20.726 20.752 70.252 60.226 10.916 10.800 10.047 120.807 20.769 10.709 20.630 20.769 10.217 80.000 30.285 10.598 30.846 80.535 10.956 20.000 30.137 80.784 10.464 50.463 100.230 80.000 10.598 20.662 60.000 40.087 20.000 10.135 10.900 10.780 100.703 10.741 10.571 20.149 90.697 30.646 10.000 30.076 10.000 10.025 70.000 30.106 40.981 10.000 10.043 50.113 30.888 10.248 120.404 30.252 40.314 10.220 50.245 10.466 60.366 10.159 20.000 30.149 50.690 20.000 10.531 20.253 10.285 40.460 10.440 40.813 10.230 10.283 40.159 90.000 10.728 10.666 40.958 10.000 10.021 40.252 40.118 30.000 50.445 30.223 90.285 10.194 30.390 20.000 10.475 10.842 70.000 10.455 30.000 10.250 40.458 70.000 10.865 10.000 10.000 10.635 10.359 20.972 10.087 20.447 10.000 10.000 50.000 10.129 20.532 50.446 60.503 30.071 110.135 110.699 30.717 10.097 10.000 10.665 10.000 20.000 21.000 10.752 40.000 20.000 10.000 10.142 80.200 10.259 11.000 10.000 1
PonderV2 ScanNet2000.346 40.552 60.270 60.175 50.810 60.682 60.950 40.560 50.641 80.761 20.398 90.357 80.570 60.113 20.804 40.603 50.750 50.283 30.681 50.952 40.548 40.874 40.852 90.290 80.700 20.356 90.792 40.445 80.545 90.436 80.351 90.787 60.611 60.050 70.290 100.519 90.000 10.825 60.888 40.842 30.259 30.100 20.558 50.070 100.497 60.247 100.457 90.889 20.248 70.106 80.817 90.691 50.094 50.729 30.636 50.620 100.503 90.660 100.243 50.000 30.212 60.590 40.860 60.400 40.881 50.000 30.202 20.622 80.408 70.499 70.261 70.000 10.385 70.636 70.000 40.000 70.000 10.000 40.433 120.843 50.660 50.574 100.481 30.336 40.677 50.486 40.000 30.030 20.000 10.034 50.000 30.080 60.869 80.000 10.000 80.000 90.540 60.727 20.232 130.115 70.186 60.193 70.000 120.403 80.326 50.103 100.000 30.290 30.392 80.000 10.346 60.062 80.424 30.375 50.431 50.667 40.115 100.082 80.239 50.000 10.504 100.606 50.584 80.000 10.002 60.186 60.104 80.000 50.394 40.384 60.083 60.000 50.007 70.000 10.000 20.880 40.000 10.377 80.000 10.263 30.565 20.000 10.608 70.000 10.000 10.304 60.009 70.924 20.000 60.000 70.000 10.000 50.000 10.128 30.584 20.475 50.412 60.076 90.269 30.621 50.509 50.010 40.000 10.491 80.063 10.000 20.472 40.880 20.000 20.000 10.000 10.179 40.125 20.000 30.441 60.000 1
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.
PTv3 ScanNet2000.393 20.592 20.330 10.216 20.851 20.687 50.971 20.586 10.755 10.752 60.505 10.404 50.575 30.000 100.848 10.616 30.761 20.349 10.738 20.978 20.546 50.860 70.926 20.346 20.654 30.384 50.828 10.523 30.699 20.583 40.387 60.822 20.688 20.118 40.474 20.603 40.000 10.832 40.903 20.753 80.140 70.000 70.650 30.109 30.520 20.457 10.497 80.871 30.281 20.192 30.887 30.748 20.168 10.727 40.733 20.740 10.644 10.714 40.190 90.000 30.256 30.449 70.914 10.514 20.759 110.337 10.172 40.692 50.617 10.636 10.325 40.000 10.641 10.782 10.000 40.065 30.000 10.000 40.842 20.903 10.661 30.662 30.612 10.405 20.731 10.566 20.000 30.000 60.000 10.017 110.301 10.088 50.941 20.000 10.077 20.000 90.717 40.790 10.310 100.026 130.264 30.349 10.220 30.397 90.366 10.115 90.000 30.337 10.463 60.000 10.531 20.218 20.593 10.455 20.469 10.708 30.210 20.592 20.108 120.000 10.728 10.682 20.671 50.000 10.000 80.407 10.136 20.022 20.575 10.436 40.259 30.428 10.048 40.000 10.000 20.879 50.000 10.480 20.000 10.133 60.597 10.000 10.690 20.000 10.000 10.009 120.000 110.921 30.000 60.151 30.000 10.000 50.000 10.109 70.494 100.622 20.394 70.073 100.141 70.798 10.528 40.026 20.000 10.551 40.000 20.000 20.134 60.717 60.000 20.000 10.000 10.188 30.000 50.000 30.791 20.000 1
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)
L3DETR-ScanNet_2000.336 60.533 90.279 40.155 60.801 80.689 30.946 50.539 70.660 60.759 30.380 100.333 100.583 20.000 100.788 70.529 80.740 60.261 80.679 70.940 90.525 90.860 70.883 50.226 90.613 70.397 40.720 90.512 40.565 80.620 20.417 40.775 90.629 40.158 20.298 80.579 80.000 10.835 20.883 50.927 10.114 80.079 40.511 80.073 90.508 40.312 40.629 40.861 40.192 120.098 110.908 20.636 90.032 130.563 130.514 110.664 40.505 80.697 60.225 70.000 30.264 20.411 90.860 60.321 90.960 10.058 20.109 100.776 20.526 30.557 20.303 60.000 10.339 80.712 40.000 40.014 50.000 10.000 40.638 80.856 30.641 60.579 90.107 130.119 110.661 70.416 50.000 30.000 60.000 10.007 130.000 30.067 80.910 40.000 10.000 80.000 90.463 70.448 50.294 120.324 10.293 20.211 60.108 60.448 70.068 130.141 50.000 30.330 20.699 10.000 10.256 70.192 40.000 110.355 60.418 60.209 130.146 80.679 10.101 130.000 10.503 110.687 10.671 50.000 10.000 80.174 70.117 40.000 50.122 60.515 20.104 40.259 20.312 30.000 10.000 20.765 90.000 10.369 100.000 10.183 50.422 100.000 10.646 30.000 10.000 10.565 20.001 100.125 130.010 40.002 60.000 10.487 10.000 10.075 120.548 30.420 70.233 120.082 70.138 100.430 100.427 90.000 90.000 10.549 50.000 20.000 20.074 70.409 120.000 20.000 10.000 10.152 60.051 30.000 30.598 40.000 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12
OA-CNN-L_ScanNet2000.333 70.558 30.269 70.124 90.821 40.703 20.946 50.569 30.662 30.748 70.487 20.455 20.572 50.000 100.789 60.534 70.736 70.271 40.713 30.949 50.498 120.877 30.860 70.332 50.706 10.474 10.788 60.406 90.637 40.495 70.355 80.805 40.592 100.015 120.396 40.602 50.000 10.799 70.876 60.713 120.276 20.000 70.493 90.080 70.448 110.363 30.661 30.833 50.262 40.125 50.823 80.665 70.076 70.720 50.557 70.637 70.517 70.672 90.227 60.000 30.158 90.496 50.843 90.352 80.835 90.000 30.103 110.711 40.527 20.526 50.320 50.000 10.568 50.625 80.067 10.000 70.000 10.001 30.806 40.836 60.621 80.591 60.373 70.314 50.668 60.398 70.003 20.000 60.000 10.016 120.024 20.043 110.906 50.000 10.052 40.000 90.384 80.330 90.342 60.100 80.223 50.183 90.112 50.476 50.313 60.130 80.196 20.112 80.370 100.000 10.234 80.071 70.160 50.403 40.398 100.492 110.197 30.076 90.272 30.000 10.200 130.560 70.735 40.000 10.000 80.000 80.110 60.002 40.021 70.412 50.000 80.000 50.000 90.000 10.000 20.794 80.000 10.445 40.000 10.022 70.509 60.000 10.517 110.000 10.000 10.001 130.245 30.915 50.024 30.089 40.000 10.262 20.000 10.103 90.524 60.392 90.515 20.013 130.251 40.411 110.662 20.001 80.000 10.473 90.000 20.000 20.150 50.699 70.000 20.000 10.000 10.166 50.000 50.024 20.000 70.000 1
PPT-SpUNet-F.T.0.332 80.556 40.270 50.123 100.816 50.682 60.946 50.549 60.657 70.756 40.459 50.376 70.550 70.001 90.807 30.616 30.727 80.267 50.691 40.942 80.530 80.872 50.874 60.330 60.542 100.374 60.792 40.400 100.673 30.572 50.433 20.793 50.623 50.008 130.351 60.594 60.000 10.783 90.876 60.833 40.213 40.000 70.537 60.091 50.519 30.304 50.620 60.942 10.264 30.124 60.855 50.695 40.086 60.646 70.506 120.658 50.535 50.715 30.314 10.000 30.241 40.608 20.897 20.359 70.858 70.000 30.076 130.611 90.392 80.509 60.378 30.000 10.579 30.565 120.000 40.000 70.000 10.000 40.755 50.806 80.661 30.572 110.350 80.181 70.660 80.300 100.000 30.000 60.000 10.023 80.000 30.042 120.930 30.000 10.000 80.077 60.584 50.392 70.339 70.185 60.171 80.308 20.006 110.563 30.256 70.150 30.000 30.002 120.345 110.000 10.045 100.197 30.063 70.323 90.453 30.600 70.163 70.037 110.349 20.000 10.672 30.679 30.753 20.000 10.000 80.000 80.117 40.000 50.000 80.291 80.000 80.000 50.039 50.000 10.000 20.899 20.000 10.374 90.000 10.000 90.545 40.000 10.634 40.000 10.000 10.074 90.223 40.914 60.000 60.021 50.000 10.000 50.000 10.112 50.498 90.649 10.383 80.095 20.135 110.449 90.432 80.008 60.000 10.518 60.000 20.000 20.000 80.796 30.000 20.000 10.000 10.138 100.000 50.000 30.000 70.000 1
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 80.806 70.670 90.943 80.535 80.662 30.705 120.423 70.407 40.505 90.003 80.765 90.582 60.686 110.227 120.680 60.943 70.601 20.854 90.892 40.335 40.417 130.357 80.724 80.453 70.632 50.596 30.432 30.783 70.512 120.021 110.244 110.637 10.000 10.787 80.873 80.743 100.000 130.000 70.534 70.110 20.499 50.289 60.626 50.620 100.168 130.204 20.849 60.679 60.117 30.633 80.684 30.650 60.552 40.684 80.312 20.000 30.175 80.429 80.865 30.413 30.837 80.000 30.145 60.626 70.451 60.487 80.513 10.000 10.529 60.613 90.000 40.033 40.000 10.000 40.828 30.871 20.622 70.587 70.411 60.137 100.645 100.343 80.000 30.000 60.000 10.022 90.000 30.026 130.829 90.000 10.022 60.089 50.842 20.253 110.318 90.296 20.178 70.291 30.224 20.584 20.200 100.132 70.000 30.128 70.227 120.000 10.230 90.047 90.149 60.331 80.412 80.618 60.164 60.102 70.522 10.000 10.655 40.378 90.469 110.000 10.000 80.000 80.105 70.000 50.000 80.483 30.000 80.000 50.028 60.000 10.000 20.906 10.000 10.339 110.000 10.000 90.457 80.000 10.612 60.000 10.000 10.408 30.000 110.900 70.000 60.000 70.000 10.029 40.000 10.074 130.455 110.479 40.427 50.079 80.140 80.496 70.414 100.022 30.000 10.471 100.000 20.000 20.000 80.722 50.000 20.000 10.000 10.138 100.000 50.000 30.000 70.000 1
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
CeCo0.340 50.551 70.247 90.181 40.784 90.661 100.939 90.564 40.624 90.721 80.484 30.429 30.575 30.027 60.774 80.503 100.753 40.242 90.656 90.945 60.534 60.865 60.860 70.177 130.616 60.400 30.818 20.579 10.615 70.367 100.408 50.726 110.633 30.162 10.360 50.619 20.000 10.828 50.873 80.924 20.109 90.083 30.564 40.057 130.475 90.266 70.781 10.767 60.257 50.100 90.825 70.663 80.048 110.620 100.551 80.595 110.532 60.692 70.246 40.000 30.213 50.615 10.861 50.376 60.900 40.000 30.102 120.660 60.321 110.547 40.226 90.000 10.311 90.742 20.011 30.006 60.000 10.000 40.546 110.824 70.345 100.665 20.450 50.435 10.683 40.411 60.338 10.000 60.000 10.030 60.000 30.068 70.892 60.000 10.063 30.000 90.257 90.304 100.387 40.079 100.228 40.190 80.000 120.586 10.347 30.133 60.000 30.037 90.377 90.000 10.384 50.006 120.003 90.421 30.410 90.643 50.171 50.121 50.142 100.000 10.510 90.447 80.474 100.000 10.000 80.286 30.083 90.000 50.000 80.603 10.096 50.063 40.000 90.000 10.000 20.898 30.000 10.429 50.000 10.400 10.550 30.000 10.633 50.000 10.000 10.377 40.000 110.916 40.000 60.000 70.000 10.000 50.000 10.102 100.499 80.296 100.463 40.089 50.304 10.740 20.401 120.010 40.000 10.560 30.000 20.000 20.709 20.652 80.000 20.000 10.000 10.143 70.000 50.000 30.609 30.000 1
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 70.782 100.634 130.937 100.489 110.578 100.721 80.364 110.355 90.515 80.023 70.764 100.523 90.707 100.264 70.633 100.922 100.507 110.886 10.804 110.179 110.436 120.300 100.656 120.529 20.501 110.394 90.296 120.820 30.603 70.131 30.179 130.619 20.000 10.707 120.865 100.773 60.171 50.010 60.484 100.063 110.463 100.254 90.332 120.649 90.220 90.100 90.729 110.613 110.071 90.582 110.628 60.702 30.424 110.749 20.137 110.000 30.142 100.360 100.863 40.305 100.877 60.000 30.173 30.606 100.337 100.478 90.154 110.000 10.253 100.664 50.000 40.000 70.000 10.000 40.626 90.782 90.302 120.602 50.185 110.282 60.651 90.317 90.000 30.000 60.000 10.022 90.000 30.154 10.876 70.000 10.014 70.063 80.029 130.553 40.467 20.084 90.124 100.157 120.049 100.373 100.252 80.097 110.000 30.219 40.542 30.000 10.392 40.172 60.000 110.339 70.417 70.533 100.093 110.115 60.195 70.000 10.516 80.288 120.741 30.000 10.001 70.233 50.056 100.000 50.159 50.334 70.077 70.000 50.000 90.000 10.000 20.749 100.000 10.411 60.000 10.008 80.452 90.000 10.595 80.000 10.000 10.220 80.006 80.894 90.006 50.000 70.000 10.000 50.000 10.112 50.504 70.404 80.551 10.093 40.129 130.484 80.381 130.000 90.000 10.396 110.000 20.000 20.620 30.402 130.000 20.000 10.000 10.142 80.000 50.000 30.512 50.000 1
LGroundpermissive0.272 110.485 110.184 110.106 110.778 110.676 80.932 110.479 130.572 110.718 100.399 80.265 110.453 120.085 30.745 110.446 110.726 90.232 110.622 110.901 110.512 100.826 110.786 120.178 120.549 80.277 110.659 110.381 110.518 100.295 130.323 100.777 80.599 80.028 90.321 70.363 120.000 10.708 110.858 110.746 90.063 110.022 50.457 110.077 80.476 80.243 110.402 100.397 130.233 80.077 130.720 130.610 120.103 40.629 90.437 130.626 90.446 100.702 50.190 90.005 10.058 120.322 110.702 120.244 110.768 100.000 30.134 90.552 110.279 120.395 110.147 120.000 10.207 110.612 100.000 40.000 70.000 10.000 40.658 70.566 110.323 110.525 130.229 100.179 80.467 130.154 120.000 30.002 40.000 10.051 10.000 30.127 20.703 100.000 10.000 80.216 10.112 120.358 80.547 10.187 50.092 120.156 130.055 80.296 110.252 80.143 40.000 30.014 100.398 70.000 10.028 120.173 50.000 110.265 120.348 110.415 120.179 40.019 120.218 60.000 10.597 60.274 130.565 90.000 10.012 50.000 80.039 120.022 20.000 80.117 110.000 80.000 50.000 90.000 10.000 20.324 120.000 10.384 70.000 10.000 90.251 130.000 10.566 90.000 10.000 10.066 100.404 10.886 100.199 10.000 70.000 10.059 30.000 10.136 10.540 40.127 130.295 90.085 60.143 60.514 60.413 110.000 90.000 10.498 70.000 20.000 20.000 80.623 90.000 20.000 10.000 10.132 120.000 50.000 30.000 70.000 1
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
CSC-Pretrainpermissive0.249 130.455 130.171 120.079 130.766 130.659 110.930 130.494 100.542 130.700 130.314 130.215 130.430 130.121 10.697 130.441 120.683 120.235 100.609 130.895 120.476 130.816 120.770 130.186 100.634 40.216 130.734 70.340 120.471 120.307 120.293 130.591 130.542 110.076 60.205 120.464 110.000 10.484 130.832 130.766 70.052 120.000 70.413 120.059 120.418 120.222 120.318 130.609 110.206 110.112 70.743 100.625 100.076 70.579 120.548 90.590 120.371 120.552 130.081 120.003 20.142 100.201 130.638 130.233 120.686 130.000 30.142 70.444 130.375 90.247 130.198 100.000 10.128 130.454 130.019 20.097 10.000 10.000 40.553 100.557 120.373 90.545 120.164 120.014 130.547 120.174 110.000 30.002 40.000 10.037 30.000 30.063 90.664 120.000 10.000 80.130 20.170 100.152 130.335 80.079 100.110 110.175 100.098 70.175 130.166 110.045 130.207 10.014 100.465 50.000 10.001 130.001 130.046 80.299 110.327 120.537 90.033 120.012 130.186 80.000 10.205 120.377 100.463 120.000 10.058 30.000 80.055 110.041 10.000 80.105 120.000 80.000 50.000 90.000 10.000 20.398 110.000 10.308 130.000 10.000 90.319 110.000 10.543 100.000 10.000 10.062 110.004 90.862 120.000 60.000 70.000 10.000 50.000 10.123 40.316 120.225 110.250 110.094 30.180 50.332 120.441 70.000 90.000 10.310 130.000 20.000 20.000 80.592 100.000 20.000 10.000 10.203 20.000 50.000 30.000 70.000 1
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
Minkowski 34Dpermissive0.253 120.463 120.154 130.102 120.771 120.650 120.932 110.483 120.571 120.710 110.331 120.250 120.492 100.044 40.703 120.419 130.606 130.227 120.621 120.865 130.531 70.771 130.813 100.291 70.484 110.242 120.612 130.282 130.440 130.351 110.299 110.622 120.593 90.027 100.293 90.310 130.000 10.757 100.858 110.737 110.150 60.164 10.368 130.084 60.381 130.142 130.357 110.720 80.214 100.092 120.724 120.596 130.056 100.655 60.525 100.581 130.352 130.594 120.056 130.000 30.014 130.224 120.772 110.205 130.720 120.000 30.159 50.531 120.163 130.294 120.136 130.000 10.169 120.589 110.000 40.000 70.000 10.002 20.663 60.466 130.265 130.582 80.337 90.016 120.559 110.084 130.000 30.000 60.000 10.036 40.000 30.125 30.670 110.000 10.102 10.071 70.164 110.406 60.386 50.046 120.068 130.159 110.117 40.284 120.111 120.094 120.000 30.000 130.197 130.000 10.044 110.013 110.002 100.228 130.307 130.588 80.025 130.545 30.134 110.000 10.655 40.302 110.282 130.000 10.060 20.000 80.035 130.000 50.000 80.097 130.000 80.000 50.005 80.000 10.000 20.096 130.000 10.334 120.000 10.000 90.274 120.000 10.513 120.000 10.000 10.280 70.194 50.897 80.000 60.000 70.000 10.000 50.000 10.108 80.279 130.189 120.141 130.059 120.272 20.307 130.445 60.003 70.000 10.353 120.000 20.026 10.000 80.581 110.001 10.000 10.000 10.093 130.002 40.000 30.000 70.000 1
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019