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 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 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
BFANet ScanNet200permissive0.360 20.553 40.293 20.193 20.827 20.689 20.970 20.528 80.661 40.753 40.436 50.378 50.469 100.042 50.810 20.654 10.760 20.266 50.659 70.973 20.574 20.849 90.897 20.382 10.546 80.372 60.698 90.491 50.617 50.526 50.436 10.764 90.476 120.101 50.409 20.585 70.000 10.835 10.901 20.810 50.102 90.000 70.688 10.096 30.483 60.264 70.612 60.591 110.358 10.161 30.863 30.707 20.128 20.814 10.669 30.629 70.563 20.651 100.258 30.000 30.194 60.494 50.806 90.394 40.953 20.000 30.233 10.757 20.508 40.556 30.476 20.000 10.573 30.741 30.000 40.000 60.000 10.000 30.000 120.852 40.678 10.616 30.460 30.338 30.710 20.534 20.000 30.025 20.000 10.043 20.000 30.056 90.493 120.000 10.000 70.109 30.785 20.590 30.298 100.282 30.143 80.262 40.053 80.526 40.337 30.215 10.000 30.135 50.510 30.000 10.596 10.043 90.511 20.321 90.459 20.772 10.124 80.060 90.266 40.000 10.574 60.568 50.653 60.000 10.093 10.298 20.239 10.000 50.516 20.129 90.284 10.000 40.431 10.000 10.000 10.848 60.000 10.492 10.000 10.376 20.522 50.000 10.469 120.000 10.000 10.330 40.151 50.875 100.000 50.254 10.000 10.000 50.000 10.088 100.661 10.481 30.255 90.105 10.139 90.666 30.641 20.000 80.000 10.614 10.000 20.000 20.000 70.921 10.000 20.000 10.000 10.497 10.000 40.000 20.000 60.000 1
PonderV2 ScanNet2000.346 30.552 50.270 50.175 40.810 50.682 50.950 30.560 40.641 70.761 10.398 80.357 70.570 50.113 20.804 40.603 40.750 40.283 20.681 40.952 30.548 30.874 30.852 80.290 70.700 20.356 80.792 30.445 70.545 80.436 70.351 80.787 50.611 50.050 70.290 90.519 90.000 10.825 50.888 30.842 30.259 20.100 20.558 40.070 90.497 50.247 90.457 80.889 20.248 60.106 70.817 80.691 40.094 50.729 20.636 40.620 90.503 80.660 90.243 50.000 30.212 50.590 30.860 60.400 30.881 40.000 30.202 20.622 70.408 60.499 70.261 70.000 10.385 60.636 60.000 40.000 60.000 10.000 30.433 110.843 50.660 40.574 90.481 20.336 40.677 40.486 30.000 30.030 10.000 10.034 50.000 30.080 50.869 70.000 10.000 70.000 80.540 50.727 20.232 120.115 60.186 50.193 60.000 110.403 70.326 40.103 90.000 30.290 30.392 70.000 10.346 50.062 70.424 30.375 40.431 40.667 30.115 90.082 70.239 50.000 10.504 90.606 40.584 70.000 10.002 50.186 50.104 70.000 50.394 30.384 60.083 50.000 40.007 60.000 10.000 10.880 40.000 10.377 70.000 10.263 30.565 20.000 10.608 60.000 10.000 10.304 50.009 60.924 10.000 50.000 60.000 10.000 50.000 10.128 20.584 20.475 50.412 50.076 90.269 30.621 40.509 40.010 30.000 10.491 70.063 10.000 20.472 30.880 20.000 20.000 10.000 10.179 40.125 10.000 20.441 50.000 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 10.592 10.330 10.216 10.851 10.687 40.971 10.586 10.755 10.752 50.505 10.404 40.575 20.000 100.848 10.616 20.761 10.349 10.738 10.978 10.546 40.860 60.926 10.346 20.654 30.384 40.828 10.523 30.699 10.583 30.387 60.822 10.688 10.118 40.474 10.603 40.000 10.832 30.903 10.753 80.140 60.000 70.650 20.109 20.520 10.457 10.497 70.871 30.281 20.192 20.887 20.748 10.168 10.727 30.733 10.740 10.644 10.714 30.190 80.000 30.256 20.449 60.914 10.514 10.759 100.337 10.172 40.692 40.617 10.636 10.325 40.000 10.641 10.782 10.000 40.065 20.000 10.000 30.842 10.903 10.661 20.662 20.612 10.405 20.731 10.566 10.000 30.000 50.000 10.017 100.301 10.088 40.941 10.000 10.077 20.000 80.717 30.790 10.310 90.026 120.264 20.349 10.220 20.397 80.366 10.115 80.000 30.337 10.463 50.000 10.531 20.218 10.593 10.455 10.469 10.708 20.210 10.592 20.108 110.000 10.728 10.682 20.671 40.000 10.000 70.407 10.136 20.022 20.575 10.436 40.259 20.428 10.048 30.000 10.000 10.879 50.000 10.480 20.000 10.133 50.597 10.000 10.690 10.000 10.000 10.009 110.000 100.921 20.000 50.151 20.000 10.000 50.000 10.109 60.494 90.622 20.394 60.073 100.141 70.798 10.528 30.026 10.000 10.551 30.000 20.000 20.134 50.717 50.000 20.000 10.000 10.188 30.000 40.000 20.791 10.000 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 50.533 80.279 30.155 50.801 70.689 20.946 40.539 60.660 50.759 20.380 90.333 90.583 10.000 100.788 60.529 70.740 50.261 70.679 60.940 80.525 80.860 60.883 40.226 80.613 60.397 30.720 80.512 40.565 70.620 10.417 40.775 80.629 30.158 20.298 70.579 80.000 10.835 10.883 40.927 10.114 70.079 40.511 70.073 80.508 30.312 30.629 30.861 40.192 110.098 100.908 10.636 80.032 120.563 120.514 100.664 30.505 70.697 50.225 70.000 30.264 10.411 80.860 60.321 80.960 10.058 20.109 90.776 10.526 30.557 20.303 60.000 10.339 70.712 40.000 40.014 40.000 10.000 30.638 70.856 30.641 50.579 80.107 120.119 100.661 60.416 40.000 30.000 50.000 10.007 120.000 30.067 70.910 30.000 10.000 70.000 80.463 60.448 50.294 110.324 10.293 10.211 50.108 50.448 60.068 120.141 40.000 30.330 20.699 10.000 10.256 60.192 30.000 100.355 50.418 50.209 120.146 70.679 10.101 120.000 10.503 100.687 10.671 40.000 10.000 70.174 60.117 30.000 50.122 50.515 20.104 30.259 20.312 20.000 10.000 10.765 80.000 10.369 90.000 10.183 40.422 90.000 10.646 20.000 10.000 10.565 10.001 90.125 120.010 30.002 50.000 10.487 10.000 10.075 110.548 30.420 60.233 110.082 70.138 100.430 90.427 80.000 80.000 10.549 40.000 20.000 20.074 60.409 110.000 20.000 10.000 10.152 60.051 20.000 20.598 30.000 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12
OA-CNN-L_ScanNet2000.333 60.558 20.269 60.124 80.821 30.703 10.946 40.569 20.662 20.748 60.487 20.455 10.572 40.000 100.789 50.534 60.736 60.271 30.713 20.949 40.498 110.877 20.860 60.332 40.706 10.474 10.788 50.406 80.637 30.495 60.355 70.805 30.592 90.015 110.396 30.602 50.000 10.799 60.876 50.713 120.276 10.000 70.493 80.080 60.448 100.363 20.661 20.833 50.262 40.125 40.823 70.665 60.076 70.720 40.557 60.637 60.517 60.672 80.227 60.000 30.158 80.496 40.843 80.352 70.835 80.000 30.103 100.711 30.527 20.526 50.320 50.000 10.568 40.625 70.067 10.000 60.000 10.001 20.806 30.836 60.621 70.591 50.373 60.314 50.668 50.398 60.003 20.000 50.000 10.016 110.024 20.043 100.906 40.000 10.052 40.000 80.384 70.330 90.342 50.100 70.223 40.183 80.112 40.476 50.313 50.130 70.196 20.112 70.370 90.000 10.234 70.071 60.160 40.403 30.398 90.492 100.197 20.076 80.272 30.000 10.200 120.560 60.735 30.000 10.000 70.000 70.110 50.002 40.021 60.412 50.000 70.000 40.000 80.000 10.000 10.794 70.000 10.445 30.000 10.022 60.509 60.000 10.517 100.000 10.000 10.001 120.245 20.915 40.024 20.089 30.000 10.262 20.000 10.103 80.524 50.392 80.515 20.013 120.251 40.411 100.662 10.001 70.000 10.473 80.000 20.000 20.150 40.699 60.000 20.000 10.000 10.166 50.000 40.024 10.000 60.000 1
PPT-SpUNet-F.T.0.332 70.556 30.270 40.123 90.816 40.682 50.946 40.549 50.657 60.756 30.459 40.376 60.550 60.001 90.807 30.616 20.727 70.267 40.691 30.942 70.530 70.872 40.874 50.330 50.542 90.374 50.792 30.400 90.673 20.572 40.433 20.793 40.623 40.008 120.351 50.594 60.000 10.783 80.876 50.833 40.213 30.000 70.537 50.091 40.519 20.304 40.620 50.942 10.264 30.124 50.855 40.695 30.086 60.646 60.506 110.658 40.535 40.715 20.314 10.000 30.241 30.608 20.897 20.359 60.858 60.000 30.076 120.611 80.392 70.509 60.378 30.000 10.579 20.565 110.000 40.000 60.000 10.000 30.755 40.806 80.661 20.572 100.350 70.181 70.660 70.300 90.000 30.000 50.000 10.023 70.000 30.042 110.930 20.000 10.000 70.077 50.584 40.392 70.339 60.185 50.171 70.308 20.006 100.563 30.256 60.150 20.000 30.002 110.345 100.000 10.045 90.197 20.063 60.323 80.453 30.600 60.163 60.037 100.349 20.000 10.672 20.679 30.753 10.000 10.000 70.000 70.117 30.000 50.000 70.291 80.000 70.000 40.039 40.000 10.000 10.899 20.000 10.374 80.000 10.000 80.545 40.000 10.634 30.000 10.000 10.074 80.223 30.914 50.000 50.021 40.000 10.000 50.000 10.112 40.498 80.649 10.383 70.095 20.135 110.449 80.432 70.008 50.000 10.518 50.000 20.000 20.000 70.796 30.000 20.000 10.000 10.138 90.000 40.000 20.000 60.000 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 80.539 70.265 70.131 70.806 60.670 80.943 70.535 70.662 20.705 110.423 60.407 30.505 80.003 80.765 80.582 50.686 100.227 110.680 50.943 60.601 10.854 80.892 30.335 30.417 120.357 70.724 70.453 60.632 40.596 20.432 30.783 60.512 110.021 100.244 100.637 10.000 10.787 70.873 70.743 100.000 120.000 70.534 60.110 10.499 40.289 50.626 40.620 90.168 120.204 10.849 50.679 50.117 30.633 70.684 20.650 50.552 30.684 70.312 20.000 30.175 70.429 70.865 30.413 20.837 70.000 30.145 60.626 60.451 50.487 80.513 10.000 10.529 50.613 80.000 40.033 30.000 10.000 30.828 20.871 20.622 60.587 60.411 50.137 90.645 90.343 70.000 30.000 50.000 10.022 80.000 30.026 120.829 80.000 10.022 50.089 40.842 10.253 110.318 80.296 20.178 60.291 30.224 10.584 20.200 90.132 60.000 30.128 60.227 110.000 10.230 80.047 80.149 50.331 70.412 70.618 50.164 50.102 60.522 10.000 10.655 30.378 80.469 100.000 10.000 70.000 70.105 60.000 50.000 70.483 30.000 70.000 40.028 50.000 10.000 10.906 10.000 10.339 100.000 10.000 80.457 70.000 10.612 50.000 10.000 10.408 20.000 100.900 60.000 50.000 60.000 10.029 40.000 10.074 120.455 100.479 40.427 40.079 80.140 80.496 60.414 90.022 20.000 10.471 90.000 20.000 20.000 70.722 40.000 20.000 10.000 10.138 90.000 40.000 20.000 60.000 1
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
CeCo0.340 40.551 60.247 80.181 30.784 80.661 90.939 80.564 30.624 80.721 70.484 30.429 20.575 20.027 60.774 70.503 90.753 30.242 80.656 80.945 50.534 50.865 50.860 60.177 120.616 50.400 20.818 20.579 10.615 60.367 90.408 50.726 100.633 20.162 10.360 40.619 20.000 10.828 40.873 70.924 20.109 80.083 30.564 30.057 120.475 80.266 60.781 10.767 60.257 50.100 80.825 60.663 70.048 110.620 90.551 70.595 100.532 50.692 60.246 40.000 30.213 40.615 10.861 50.376 50.900 30.000 30.102 110.660 50.321 100.547 40.226 80.000 10.311 80.742 20.011 30.006 50.000 10.000 30.546 100.824 70.345 90.665 10.450 40.435 10.683 30.411 50.338 10.000 50.000 10.030 60.000 30.068 60.892 50.000 10.063 30.000 80.257 80.304 100.387 30.079 90.228 30.190 70.000 110.586 10.347 20.133 50.000 30.037 80.377 80.000 10.384 40.006 110.003 80.421 20.410 80.643 40.171 40.121 40.142 90.000 10.510 80.447 70.474 90.000 10.000 70.286 30.083 80.000 50.000 70.603 10.096 40.063 30.000 80.000 10.000 10.898 30.000 10.429 40.000 10.400 10.550 30.000 10.633 40.000 10.000 10.377 30.000 100.916 30.000 50.000 60.000 10.000 50.000 10.102 90.499 70.296 90.463 30.089 50.304 10.740 20.401 110.010 30.000 10.560 20.000 20.000 20.709 10.652 70.000 20.000 10.000 10.143 70.000 40.000 20.609 20.000 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 90.508 90.225 90.142 60.782 90.634 120.937 90.489 100.578 90.721 70.364 100.355 80.515 70.023 70.764 90.523 80.707 90.264 60.633 90.922 90.507 100.886 10.804 100.179 100.436 110.300 90.656 110.529 20.501 100.394 80.296 110.820 20.603 60.131 30.179 120.619 20.000 10.707 110.865 90.773 60.171 40.010 60.484 90.063 100.463 90.254 80.332 110.649 80.220 80.100 80.729 100.613 100.071 90.582 100.628 50.702 20.424 100.749 10.137 100.000 30.142 90.360 90.863 40.305 90.877 50.000 30.173 30.606 90.337 90.478 90.154 100.000 10.253 90.664 50.000 40.000 60.000 10.000 30.626 80.782 90.302 110.602 40.185 100.282 60.651 80.317 80.000 30.000 50.000 10.022 80.000 30.154 10.876 60.000 10.014 60.063 70.029 120.553 40.467 20.084 80.124 90.157 110.049 90.373 90.252 70.097 100.000 30.219 40.542 20.000 10.392 30.172 50.000 100.339 60.417 60.533 90.093 100.115 50.195 70.000 10.516 70.288 110.741 20.000 10.001 60.233 40.056 90.000 50.159 40.334 70.077 60.000 40.000 80.000 10.000 10.749 90.000 10.411 50.000 10.008 70.452 80.000 10.595 70.000 10.000 10.220 70.006 70.894 80.006 40.000 60.000 10.000 50.000 10.112 40.504 60.404 70.551 10.093 40.129 120.484 70.381 120.000 80.000 10.396 100.000 20.000 20.620 20.402 120.000 20.000 10.000 10.142 80.000 40.000 20.512 40.000 1
LGroundpermissive0.272 100.485 100.184 100.106 100.778 100.676 70.932 100.479 120.572 100.718 90.399 70.265 100.453 110.085 30.745 100.446 100.726 80.232 100.622 100.901 100.512 90.826 100.786 110.178 110.549 70.277 100.659 100.381 100.518 90.295 120.323 90.777 70.599 70.028 80.321 60.363 110.000 10.708 100.858 100.746 90.063 100.022 50.457 100.077 70.476 70.243 100.402 90.397 120.233 70.077 120.720 120.610 110.103 40.629 80.437 120.626 80.446 90.702 40.190 80.005 10.058 110.322 100.702 110.244 100.768 90.000 30.134 80.552 100.279 110.395 100.147 110.000 10.207 100.612 90.000 40.000 60.000 10.000 30.658 60.566 100.323 100.525 120.229 90.179 80.467 120.154 110.000 30.002 30.000 10.051 10.000 30.127 20.703 90.000 10.000 70.216 10.112 110.358 80.547 10.187 40.092 110.156 120.055 70.296 100.252 70.143 30.000 30.014 90.398 60.000 10.028 110.173 40.000 100.265 110.348 100.415 110.179 30.019 110.218 60.000 10.597 50.274 120.565 80.000 10.012 40.000 70.039 110.022 20.000 70.117 100.000 70.000 40.000 80.000 10.000 10.324 110.000 10.384 60.000 10.000 80.251 120.000 10.566 80.000 10.000 10.066 90.404 10.886 90.199 10.000 60.000 10.059 30.000 10.136 10.540 40.127 120.295 80.085 60.143 60.514 50.413 100.000 80.000 10.498 60.000 20.000 20.000 70.623 80.000 20.000 10.000 10.132 110.000 40.000 20.000 60.000 1
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
CSC-Pretrainpermissive0.249 120.455 120.171 110.079 120.766 120.659 100.930 120.494 90.542 120.700 120.314 120.215 120.430 120.121 10.697 120.441 110.683 110.235 90.609 120.895 110.476 120.816 110.770 120.186 90.634 40.216 120.734 60.340 110.471 110.307 110.293 120.591 120.542 100.076 60.205 110.464 100.000 10.484 120.832 120.766 70.052 110.000 70.413 110.059 110.418 110.222 110.318 120.609 100.206 100.112 60.743 90.625 90.076 70.579 110.548 80.590 110.371 110.552 120.081 110.003 20.142 90.201 120.638 120.233 110.686 120.000 30.142 70.444 120.375 80.247 120.198 90.000 10.128 120.454 120.019 20.097 10.000 10.000 30.553 90.557 110.373 80.545 110.164 110.014 120.547 110.174 100.000 30.002 30.000 10.037 30.000 30.063 80.664 110.000 10.000 70.130 20.170 90.152 120.335 70.079 90.110 100.175 90.098 60.175 120.166 100.045 120.207 10.014 90.465 40.000 10.001 120.001 120.046 70.299 100.327 110.537 80.033 110.012 120.186 80.000 10.205 110.377 90.463 110.000 10.058 30.000 70.055 100.041 10.000 70.105 110.000 70.000 40.000 80.000 10.000 10.398 100.000 10.308 120.000 10.000 80.319 100.000 10.543 90.000 10.000 10.062 100.004 80.862 110.000 50.000 60.000 10.000 50.000 10.123 30.316 110.225 100.250 100.094 30.180 50.332 110.441 60.000 80.000 10.310 120.000 20.000 20.000 70.592 90.000 20.000 10.000 10.203 20.000 40.000 20.000 60.000 1
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
Minkowski 34Dpermissive0.253 110.463 110.154 120.102 110.771 110.650 110.932 100.483 110.571 110.710 100.331 110.250 110.492 90.044 40.703 110.419 120.606 120.227 110.621 110.865 120.531 60.771 120.813 90.291 60.484 100.242 110.612 120.282 120.440 120.351 100.299 100.622 110.593 80.027 90.293 80.310 120.000 10.757 90.858 100.737 110.150 50.164 10.368 120.084 50.381 120.142 120.357 100.720 70.214 90.092 110.724 110.596 120.056 100.655 50.525 90.581 120.352 120.594 110.056 120.000 30.014 120.224 110.772 100.205 120.720 110.000 30.159 50.531 110.163 120.294 110.136 120.000 10.169 110.589 100.000 40.000 60.000 10.002 10.663 50.466 120.265 120.582 70.337 80.016 110.559 100.084 120.000 30.000 50.000 10.036 40.000 30.125 30.670 100.000 10.102 10.071 60.164 100.406 60.386 40.046 110.068 120.159 100.117 30.284 110.111 110.094 110.000 30.000 120.197 120.000 10.044 100.013 100.002 90.228 120.307 120.588 70.025 120.545 30.134 100.000 10.655 30.302 100.282 120.000 10.060 20.000 70.035 120.000 50.000 70.097 120.000 70.000 40.005 70.000 10.000 10.096 120.000 10.334 110.000 10.000 80.274 110.000 10.513 110.000 10.000 10.280 60.194 40.897 70.000 50.000 60.000 10.000 50.000 10.108 70.279 120.189 110.141 120.059 110.272 20.307 120.445 50.003 60.000 10.353 110.000 20.026 10.000 70.581 100.001 10.000 10.000 10.093 120.002 30.000 20.000 60.000 1
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 ap 25%head ap 25%common ap 25%tail ap 25%chairtabledoorcouchcabinetshelfdeskoffice 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 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
TD3D Scannet200permissive0.379 20.603 20.306 20.190 20.885 10.755 10.800 20.958 10.390 20.260 20.866 20.232 10.979 20.523 30.869 30.559 50.689 21.000 10.795 10.905 20.748 10.173 50.825 10.173 20.970 10.457 10.615 20.456 20.200 10.621 40.906 20.553 10.517 10.510 10.220 20.715 10.706 21.000 10.113 20.792 10.717 20.073 20.635 20.557 10.638 11.000 10.205 50.146 31.000 10.769 50.186 21.000 10.710 50.778 10.415 10.834 40.226 20.021 20.590 20.356 20.817 10.477 51.000 10.000 10.635 10.843 20.427 10.270 40.125 20.000 20.102 31.000 10.125 20.000 20.000 10.000 20.000 30.125 40.370 30.622 50.221 10.196 20.836 10.288 20.000 20.093 20.020 20.294 20.000 10.075 20.667 10.038 10.111 10.250 40.000 40.526 20.495 30.908 10.111 30.259 10.003 30.667 20.045 50.000 20.000 10.400 10.274 30.000 10.274 20.226 20.000 10.520 20.302 50.731 20.103 30.458 10.500 10.000 11.000 10.472 10.792 30.000 10.088 20.061 20.250 10.009 20.250 20.333 30.181 20.396 20.051 20.012 10.000 10.458 40.000 10.424 50.000 10.101 20.390 50.000 10.833 20.000 10.000 10.857 20.222 31.000 10.000 10.003 20.000 10.000 20.000 10.102 20.275 50.400 20.735 20.061 30.433 30.533 30.625 10.000 20.000 10.259 40.000 10.000 10.000 20.500 20.000 10.000 21.000 10.600 10.000 20.250 10.000 20.000 1
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
Mask3D Scannet2000.445 10.653 10.392 10.254 10.844 20.746 20.818 10.888 40.556 10.262 10.890 10.025 21.000 10.608 10.930 10.694 30.721 10.930 50.686 30.966 10.615 40.440 10.725 40.201 10.890 30.414 40.827 10.552 10.158 50.806 10.924 10.042 30.512 20.412 50.226 10.604 30.830 11.000 10.125 10.792 10.815 10.097 10.648 10.551 20.354 41.000 10.630 10.241 21.000 10.853 10.204 10.974 40.841 10.778 10.358 20.927 10.300 10.045 10.640 10.363 10.745 20.710 11.000 10.000 10.330 20.943 10.315 20.600 11.000 10.027 10.080 50.556 50.500 10.409 10.000 10.194 11.000 10.500 10.493 20.761 20.053 40.042 30.780 20.454 10.009 10.333 10.050 10.321 10.000 10.084 10.552 20.008 20.027 20.750 10.500 10.442 30.657 10.765 20.120 20.183 30.021 21.000 10.510 20.016 10.000 10.400 10.619 10.000 10.396 10.290 10.000 10.741 10.699 11.000 10.260 10.017 30.125 50.000 10.792 40.399 41.000 10.000 10.049 30.265 10.063 30.000 31.000 10.335 20.381 10.500 10.250 10.004 20.000 10.727 20.000 10.538 30.000 10.188 10.677 20.000 10.930 10.000 10.000 10.966 10.391 10.908 20.000 10.028 10.000 11.000 10.000 10.152 10.451 20.458 10.971 10.573 10.606 10.167 50.625 10.004 10.000 10.058 50.000 10.000 11.000 11.000 10.000 10.056 10.000 20.200 30.309 10.000 21.000 10.000 1
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
Minkowski 34D Inst.permissive0.280 40.488 40.192 50.124 40.804 40.518 40.772 50.904 30.337 50.191 40.443 40.000 30.861 40.502 40.868 40.669 40.587 40.997 30.467 50.828 50.732 20.342 30.745 30.119 50.918 20.404 50.419 40.398 30.172 30.618 50.743 40.167 20.077 50.500 20.000 30.568 40.506 51.000 10.044 40.000 30.502 40.010 40.593 40.284 50.305 50.903 50.213 40.142 40.981 30.790 40.000 41.000 10.715 40.538 50.346 40.830 50.067 30.000 30.400 30.074 40.333 40.551 21.000 10.000 10.292 30.777 40.118 50.317 30.100 40.000 20.191 20.648 30.000 30.000 20.000 10.000 20.000 30.500 10.213 50.825 10.021 50.333 10.648 50.098 40.000 20.000 30.000 30.077 30.000 10.000 50.150 50.000 30.000 30.000 50.225 20.281 40.447 40.000 50.090 40.148 40.000 40.479 50.542 10.000 20.000 10.200 30.131 50.000 10.250 30.000 40.000 10.159 50.396 40.677 30.021 40.000 40.500 10.000 11.000 10.442 30.125 50.000 10.000 40.000 30.000 40.333 10.000 30.528 10.000 30.000 30.000 30.000 30.000 10.200 50.000 10.516 40.000 10.000 30.500 30.000 10.833 20.000 10.000 10.286 40.083 40.750 30.000 10.000 30.000 10.000 20.000 10.059 50.445 30.200 30.535 40.070 20.167 40.385 40.375 30.000 20.000 10.333 30.000 10.000 10.000 20.500 20.000 10.000 20.000 20.200 30.000 20.000 20.000 20.000 1
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
CSC-Pretrain Inst.permissive0.275 50.466 50.218 40.110 50.783 50.383 50.783 40.829 50.367 40.168 50.305 50.000 30.661 50.413 50.869 20.719 10.546 50.997 30.685 40.841 40.555 50.277 40.768 20.132 30.779 50.448 30.364 50.212 50.161 40.768 20.692 50.000 40.395 30.500 20.000 30.450 50.591 31.000 10.020 50.000 30.423 50.007 50.625 30.420 30.505 31.000 10.353 20.119 50.571 40.819 20.014 31.000 10.774 20.689 40.311 50.866 20.067 30.000 30.400 30.000 50.278 50.501 31.000 10.000 10.162 50.584 50.286 30.206 50.125 20.000 20.084 40.649 20.000 30.000 20.000 10.000 20.000 30.125 40.312 40.727 30.221 20.000 40.667 40.114 30.000 20.000 30.000 30.065 50.000 10.004 40.278 30.000 30.000 30.500 20.000 40.571 10.000 50.250 40.019 50.145 50.000 40.667 20.200 40.000 20.000 10.200 30.258 40.000 10.000 40.000 40.000 10.369 40.429 30.613 40.000 50.000 40.500 10.000 10.500 50.333 50.500 40.000 10.106 10.000 30.000 40.000 30.000 30.333 30.000 30.000 30.000 30.000 30.000 10.918 10.000 10.638 10.000 10.000 30.750 10.000 10.833 20.000 10.000 10.143 50.000 50.750 30.000 10.000 30.000 10.000 20.000 10.063 40.377 40.200 30.222 50.055 40.500 20.677 20.250 40.000 20.000 10.500 20.000 10.000 10.000 20.500 20.000 10.000 20.000 20.115 50.000 20.000 20.000 20.000 1
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
LGround Inst.permissive0.314 30.529 30.225 30.155 30.810 30.625 30.798 30.940 20.372 30.217 30.484 30.000 30.927 30.528 20.826 50.694 20.605 31.000 10.731 20.846 30.716 30.350 20.589 50.123 40.857 40.457 20.578 30.376 40.183 20.765 30.800 30.000 40.278 40.500 20.000 30.659 20.569 41.000 10.093 30.000 30.539 30.010 30.578 50.378 40.571 21.000 10.337 30.252 10.530 50.814 30.000 40.744 50.743 30.746 30.346 30.863 30.067 30.000 30.400 30.167 30.667 30.488 41.000 10.000 10.208 40.783 30.166 40.375 20.071 50.000 20.200 10.607 40.000 30.000 20.000 10.000 21.000 10.500 10.517 10.716 40.221 20.000 40.706 30.085 50.000 20.000 30.000 30.077 40.000 10.063 30.278 30.000 30.000 30.500 20.083 30.181 50.515 20.286 30.144 10.219 20.042 10.582 40.400 30.000 20.000 10.000 50.305 20.000 10.000 40.036 30.000 10.413 30.500 20.533 50.250 20.200 20.500 10.000 11.000 10.472 11.000 10.000 10.000 40.000 30.250 10.000 30.000 30.333 30.000 30.000 30.000 30.000 30.000 10.600 30.000 10.594 20.000 10.000 30.500 30.000 10.647 50.000 10.000 10.429 30.333 20.500 50.000 10.000 30.000 10.000 20.000 10.069 30.696 10.050 50.556 30.031 50.042 50.750 10.250 40.000 20.000 10.630 10.000 10.000 10.000 20.500 20.000 10.000 20.000 20.400 20.000 20.000 20.000 20.000 1
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 ScanNet0.794 10.941 30.813 190.851 70.782 60.890 20.597 10.916 20.696 80.713 30.979 10.635 10.384 20.793 20.907 80.821 40.790 310.696 110.967 30.903 10.805 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)
PonderV20.785 20.978 10.800 270.833 230.788 40.853 160.545 160.910 50.713 10.705 40.979 10.596 70.390 10.769 120.832 410.821 40.792 300.730 10.975 10.897 40.785 4
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 30.964 20.855 10.843 160.781 70.858 120.575 60.831 330.685 140.714 20.979 10.594 80.310 260.801 10.892 160.841 20.819 40.723 40.940 130.887 60.725 24
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
Swin3Dpermissive0.779 40.861 200.818 140.836 200.790 30.875 40.576 50.905 60.704 50.739 10.969 100.611 20.349 100.756 220.958 10.702 450.805 140.708 70.916 320.898 30.801 2
TTT-KD0.773 50.646 910.818 140.809 350.774 90.878 30.581 20.943 10.687 120.704 50.978 40.607 50.336 150.775 80.912 60.838 30.823 20.694 120.967 30.899 20.794 3
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 60.939 40.824 60.854 60.771 100.840 300.564 100.900 80.686 130.677 120.961 160.537 310.348 110.769 120.903 100.785 100.815 60.676 220.939 140.880 110.772 8
OctFormerpermissive0.766 70.925 70.808 230.849 90.786 50.846 260.566 90.876 150.690 100.674 140.960 170.576 170.226 670.753 240.904 90.777 120.815 60.722 50.923 270.877 130.776 7
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
PPT-SpUNet-Joint0.766 70.932 50.794 330.829 250.751 220.854 140.540 200.903 70.630 340.672 150.963 140.565 210.357 80.788 30.900 120.737 250.802 150.685 170.950 70.887 60.780 5
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
OccuSeg+Semantic0.764 90.758 580.796 310.839 180.746 250.907 10.562 110.850 250.680 160.672 150.978 40.610 30.335 170.777 60.819 450.847 10.830 10.691 140.972 20.885 80.727 22
CU-Hybrid Net0.764 90.924 80.819 120.840 170.757 170.853 160.580 30.848 260.709 30.643 240.958 210.587 120.295 330.753 240.884 200.758 190.815 60.725 30.927 240.867 220.743 15
O-CNNpermissive0.762 110.924 80.823 70.844 150.770 110.852 180.577 40.847 280.711 20.640 280.958 210.592 90.217 730.762 170.888 170.758 190.813 100.726 20.932 220.868 210.744 14
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
DTC0.757 120.843 260.820 100.847 120.791 20.862 100.511 330.870 170.707 40.652 200.954 350.604 60.279 440.760 180.942 20.734 260.766 440.701 100.884 540.874 190.736 16
OA-CNN-L_ScanNet200.756 130.783 440.826 50.858 40.776 80.837 330.548 150.896 110.649 260.675 130.962 150.586 130.335 170.771 110.802 490.770 150.787 330.691 140.936 170.880 110.761 10
PNE0.755 140.786 420.835 40.834 220.758 150.849 210.570 80.836 320.648 270.668 170.978 40.581 160.367 60.683 350.856 290.804 60.801 190.678 190.961 50.889 50.716 29
P. Hermosilla: Point Neighborhood Embeddings.
ConDaFormer0.755 140.927 60.822 80.836 200.801 10.849 210.516 300.864 220.651 250.680 110.958 210.584 150.282 410.759 200.855 310.728 280.802 150.678 190.880 590.873 200.756 12
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
DMF-Net0.752 160.906 120.793 350.802 410.689 400.825 460.556 120.867 180.681 150.602 440.960 170.555 270.365 70.779 50.859 260.747 220.795 270.717 60.917 310.856 300.764 9
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
PointTransformerV20.752 160.742 660.809 220.872 10.758 150.860 110.552 130.891 130.610 410.687 60.960 170.559 250.304 290.766 150.926 40.767 160.797 230.644 330.942 110.876 160.722 26
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
PointConvFormer0.749 180.793 400.790 360.807 370.750 240.856 130.524 260.881 140.588 530.642 270.977 80.591 100.274 470.781 40.929 30.804 60.796 240.642 340.947 90.885 80.715 30
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
BPNetcopyleft0.749 180.909 100.818 140.811 330.752 200.839 320.485 470.842 290.673 180.644 230.957 250.528 370.305 280.773 90.859 260.788 80.818 50.693 130.916 320.856 300.723 25
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
MSP0.748 200.623 940.804 250.859 30.745 260.824 480.501 370.912 40.690 100.685 80.956 260.567 200.320 230.768 140.918 50.720 330.802 150.676 220.921 290.881 100.779 6
StratifiedFormerpermissive0.747 210.901 130.803 260.845 140.757 170.846 260.512 320.825 360.696 80.645 220.956 260.576 170.262 580.744 290.861 250.742 230.770 420.705 80.899 440.860 270.734 17
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
VMNetpermissive0.746 220.870 180.838 20.858 40.729 310.850 200.501 370.874 160.587 540.658 190.956 260.564 220.299 310.765 160.900 120.716 360.812 110.631 390.939 140.858 280.709 31
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)
Virtual MVFusion0.746 220.771 520.819 120.848 110.702 370.865 90.397 850.899 90.699 60.664 180.948 550.588 110.330 190.746 280.851 350.764 170.796 240.704 90.935 180.866 230.728 20
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
DiffSeg3D20.745 240.725 750.814 180.837 190.751 220.831 400.514 310.896 110.674 170.684 90.960 170.564 220.303 300.773 90.820 440.713 390.798 220.690 160.923 270.875 170.757 11
Retro-FPN0.744 250.842 270.800 270.767 550.740 270.836 350.541 180.914 30.672 190.626 320.958 210.552 280.272 490.777 60.886 190.696 460.801 190.674 250.941 120.858 280.717 27
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-Net0.743 260.620 950.799 300.849 90.730 300.822 500.493 440.897 100.664 200.681 100.955 290.562 240.378 30.760 180.903 100.738 240.801 190.673 260.907 360.877 130.745 13
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
LRPNet0.742 270.816 350.806 240.807 370.752 200.828 440.575 60.839 310.699 60.637 290.954 350.520 400.320 230.755 230.834 390.760 180.772 390.676 220.915 340.862 250.717 27
SAT0.742 270.860 210.765 490.819 280.769 120.848 230.533 220.829 340.663 210.631 310.955 290.586 130.274 470.753 240.896 140.729 270.760 500.666 280.921 290.855 320.733 18
LargeKernel3D0.739 290.909 100.820 100.806 390.740 270.852 180.545 160.826 350.594 520.643 240.955 290.541 300.263 570.723 330.858 280.775 140.767 430.678 190.933 200.848 370.694 36
Yukang Chen*, Jianhui Liu*, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs. CVPR 2023
RPN0.736 300.776 480.790 360.851 70.754 190.854 140.491 460.866 200.596 510.686 70.955 290.536 320.342 130.624 500.869 220.787 90.802 150.628 400.927 240.875 170.704 33
MinkowskiNetpermissive0.736 300.859 220.818 140.832 240.709 350.840 300.521 280.853 240.660 230.643 240.951 450.544 290.286 390.731 310.893 150.675 550.772 390.683 180.874 660.852 350.727 22
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
IPCA0.731 320.890 140.837 30.864 20.726 320.873 50.530 250.824 370.489 870.647 210.978 40.609 40.336 150.624 500.733 580.758 190.776 370.570 650.949 80.877 130.728 20
PointTransformer++0.725 330.727 740.811 210.819 280.765 130.841 290.502 360.814 420.621 370.623 340.955 290.556 260.284 400.620 520.866 230.781 110.757 540.648 310.932 220.862 250.709 31
SparseConvNet0.725 330.647 900.821 90.846 130.721 330.869 60.533 220.754 580.603 470.614 360.955 290.572 190.325 210.710 340.870 210.724 310.823 20.628 400.934 190.865 240.683 39
MatchingNet0.724 350.812 370.812 200.810 340.735 290.834 370.495 430.860 230.572 610.602 440.954 350.512 420.280 430.757 210.845 370.725 300.780 350.606 500.937 160.851 360.700 35
INS-Conv-semantic0.717 360.751 610.759 520.812 320.704 360.868 70.537 210.842 290.609 430.608 400.953 390.534 340.293 340.616 530.864 240.719 350.793 280.640 350.933 200.845 410.663 45
PointMetaBase0.714 370.835 280.785 380.821 260.684 420.846 260.531 240.865 210.614 380.596 480.953 390.500 450.246 630.674 360.888 170.692 470.764 460.624 420.849 810.844 420.675 41
contrastBoundarypermissive0.705 380.769 550.775 430.809 350.687 410.820 530.439 730.812 430.661 220.591 500.945 630.515 410.171 910.633 470.856 290.720 330.796 240.668 270.889 510.847 380.689 37
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
ClickSeg_Semantic0.703 390.774 500.800 270.793 460.760 140.847 250.471 510.802 460.463 940.634 300.968 120.491 480.271 510.726 320.910 70.706 410.815 60.551 770.878 600.833 430.570 77
RFCR0.702 400.889 150.745 630.813 310.672 450.818 570.493 440.815 410.623 350.610 380.947 570.470 570.249 620.594 560.848 360.705 420.779 360.646 320.892 490.823 490.611 60
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 Click0.701 410.825 320.796 310.723 620.716 340.832 390.433 750.816 390.634 320.609 390.969 100.418 830.344 120.559 680.833 400.715 370.808 130.560 710.902 410.847 380.680 40
JSENetpermissive0.699 420.881 170.762 500.821 260.667 460.800 690.522 270.792 490.613 390.607 410.935 830.492 470.205 780.576 610.853 330.691 490.758 520.652 300.872 690.828 460.649 49
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-Click0.693 430.743 650.794 330.655 850.684 420.822 500.497 420.719 680.622 360.617 350.977 80.447 700.339 140.750 270.664 740.703 440.790 310.596 550.946 100.855 320.647 50
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
PicassoNet-IIpermissive0.692 440.732 700.772 440.786 470.677 440.866 80.517 290.848 260.509 800.626 320.952 430.536 320.225 690.545 740.704 650.689 520.810 120.564 700.903 400.854 340.729 19
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
Feature_GeometricNetpermissive0.690 450.884 160.754 560.795 440.647 530.818 570.422 770.802 460.612 400.604 420.945 630.462 600.189 860.563 670.853 330.726 290.765 450.632 380.904 380.821 520.606 64
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
FusionNet0.688 460.704 800.741 670.754 590.656 480.829 420.501 370.741 630.609 430.548 580.950 490.522 390.371 40.633 470.756 530.715 370.771 410.623 430.861 770.814 550.658 46
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
Feature-Geometry Netpermissive0.685 470.866 190.748 600.819 280.645 550.794 720.450 630.802 460.587 540.604 420.945 630.464 590.201 810.554 700.840 380.723 320.732 640.602 530.907 360.822 510.603 67
KP-FCNN0.684 480.847 250.758 540.784 490.647 530.814 600.473 500.772 520.605 450.594 490.935 830.450 680.181 890.587 570.805 480.690 500.785 340.614 460.882 560.819 530.632 56
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
VACNN++0.684 480.728 730.757 550.776 520.690 380.804 670.464 560.816 390.577 600.587 510.945 630.508 440.276 460.671 370.710 630.663 600.750 580.589 600.881 570.832 450.653 48
DGNet0.684 480.712 790.784 390.782 510.658 470.835 360.499 410.823 380.641 290.597 470.950 490.487 500.281 420.575 620.619 780.647 680.764 460.620 450.871 720.846 400.688 38
PointContrast_LA_SEM0.683 510.757 590.784 390.786 470.639 570.824 480.408 800.775 510.604 460.541 600.934 870.532 350.269 530.552 710.777 510.645 710.793 280.640 350.913 350.824 480.671 42
Superpoint Network0.683 510.851 240.728 710.800 430.653 500.806 650.468 530.804 440.572 610.602 440.946 600.453 670.239 660.519 790.822 420.689 520.762 490.595 570.895 470.827 470.630 57
VI-PointConv0.676 530.770 540.754 560.783 500.621 610.814 600.552 130.758 560.571 630.557 560.954 350.529 360.268 550.530 770.682 690.675 550.719 670.603 520.888 520.833 430.665 44
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
ROSMRF3D0.673 540.789 410.748 600.763 570.635 590.814 600.407 820.747 600.581 580.573 530.950 490.484 510.271 510.607 540.754 540.649 650.774 380.596 550.883 550.823 490.606 64
SALANet0.670 550.816 350.770 470.768 540.652 510.807 640.451 600.747 600.659 240.545 590.924 930.473 560.149 1010.571 640.811 470.635 740.746 590.623 430.892 490.794 680.570 77
O3DSeg0.668 560.822 330.771 460.496 1050.651 520.833 380.541 180.761 550.555 690.611 370.966 130.489 490.370 50.388 990.580 810.776 130.751 560.570 650.956 60.817 540.646 51
PointConvpermissive0.666 570.781 450.759 520.699 700.644 560.822 500.475 490.779 500.564 660.504 760.953 390.428 770.203 800.586 590.754 540.661 610.753 550.588 610.902 410.813 570.642 52
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointASNLpermissive0.666 570.703 810.781 410.751 610.655 490.830 410.471 510.769 530.474 900.537 620.951 450.475 550.279 440.635 450.698 680.675 550.751 560.553 760.816 880.806 590.703 34
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PPCNN++permissive0.663 590.746 630.708 740.722 630.638 580.820 530.451 600.566 960.599 490.541 600.950 490.510 430.313 250.648 420.819 450.616 790.682 820.590 590.869 730.810 580.656 47
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
MVF-GNN0.658 600.558 1020.751 580.655 850.690 380.722 940.453 590.867 180.579 590.576 520.893 1050.523 380.293 340.733 300.571 830.692 470.659 890.606 500.875 630.804 610.668 43
DCM-Net0.658 600.778 460.702 770.806 390.619 620.813 630.468 530.693 760.494 830.524 680.941 750.449 690.298 320.510 810.821 430.675 550.727 660.568 680.826 860.803 620.637 54
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
HPGCNN0.656 620.698 830.743 650.650 870.564 790.820 530.505 350.758 560.631 330.479 800.945 630.480 530.226 670.572 630.774 520.690 500.735 620.614 460.853 800.776 830.597 70
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
SAFNet-segpermissive0.654 630.752 600.734 690.664 830.583 740.815 590.399 840.754 580.639 300.535 640.942 730.470 570.309 270.665 380.539 850.650 640.708 720.635 370.857 790.793 700.642 52
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Netpermissive0.645 640.778 460.731 700.699 700.577 750.829 420.446 650.736 640.477 890.523 700.945 630.454 640.269 530.484 890.749 570.618 770.738 600.599 540.827 850.792 730.621 59
PointConv-SFPN0.641 650.776 480.703 760.721 640.557 820.826 450.451 600.672 810.563 670.483 790.943 720.425 800.162 960.644 430.726 590.659 620.709 710.572 640.875 630.786 780.559 83
MVPNetpermissive0.641 650.831 290.715 720.671 800.590 700.781 780.394 860.679 780.642 280.553 570.937 800.462 600.256 590.649 410.406 990.626 750.691 790.666 280.877 610.792 730.608 63
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
PointMRNet0.640 670.717 780.701 780.692 730.576 760.801 680.467 550.716 690.563 670.459 860.953 390.429 760.169 930.581 600.854 320.605 800.710 690.550 780.894 480.793 700.575 75
FPConvpermissive0.639 680.785 430.760 510.713 680.603 650.798 700.392 870.534 1010.603 470.524 680.948 550.457 620.250 610.538 750.723 610.598 840.696 770.614 460.872 690.799 630.567 80
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
PD-Net0.638 690.797 390.769 480.641 930.590 700.820 530.461 570.537 1000.637 310.536 630.947 570.388 900.206 770.656 390.668 720.647 680.732 640.585 620.868 740.793 700.473 103
PointSPNet0.637 700.734 690.692 850.714 670.576 760.797 710.446 650.743 620.598 500.437 910.942 730.403 860.150 1000.626 490.800 500.649 650.697 760.557 740.846 820.777 820.563 81
SConv0.636 710.830 300.697 810.752 600.572 780.780 800.445 670.716 690.529 730.530 650.951 450.446 710.170 920.507 840.666 730.636 730.682 820.541 840.886 530.799 630.594 71
Supervoxel-CNN0.635 720.656 880.711 730.719 650.613 630.757 890.444 700.765 540.534 720.566 540.928 910.478 540.272 490.636 440.531 870.664 590.645 930.508 910.864 760.792 730.611 60
joint point-basedpermissive0.634 730.614 960.778 420.667 820.633 600.825 460.420 780.804 440.467 920.561 550.951 450.494 460.291 360.566 650.458 940.579 900.764 460.559 730.838 830.814 550.598 69
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
PointMTL0.632 740.731 710.688 880.675 770.591 690.784 770.444 700.565 970.610 410.492 770.949 530.456 630.254 600.587 570.706 640.599 830.665 880.612 490.868 740.791 760.579 74
3DSM_DMMF0.631 750.626 930.745 630.801 420.607 640.751 900.506 340.729 670.565 650.491 780.866 1080.434 720.197 840.595 550.630 770.709 400.705 740.560 710.875 630.740 930.491 98
PointNet2-SFPN0.631 750.771 520.692 850.672 780.524 870.837 330.440 720.706 740.538 710.446 880.944 690.421 820.219 720.552 710.751 560.591 860.737 610.543 830.901 430.768 850.557 84
APCF-Net0.631 750.742 660.687 900.672 780.557 820.792 750.408 800.665 820.545 700.508 730.952 430.428 770.186 870.634 460.702 660.620 760.706 730.555 750.873 670.798 650.581 73
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
FusionAwareConv0.630 780.604 980.741 670.766 560.590 700.747 910.501 370.734 650.503 820.527 660.919 970.454 640.323 220.550 730.420 980.678 540.688 800.544 810.896 460.795 670.627 58
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DenSeR0.628 790.800 380.625 1010.719 650.545 840.806 650.445 670.597 900.448 970.519 710.938 790.481 520.328 200.489 880.499 920.657 630.759 510.592 580.881 570.797 660.634 55
SegGroup_sempermissive0.627 800.818 340.747 620.701 690.602 660.764 860.385 910.629 870.490 850.508 730.931 900.409 850.201 810.564 660.725 600.618 770.692 780.539 850.873 670.794 680.548 87
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
SIConv0.625 810.830 300.694 830.757 580.563 800.772 840.448 640.647 850.520 760.509 720.949 530.431 750.191 850.496 860.614 790.647 680.672 860.535 870.876 620.783 790.571 76
dtc_net0.625 810.703 810.751 580.794 450.535 850.848 230.480 480.676 800.528 740.469 830.944 690.454 640.004 1140.464 910.636 760.704 430.758 520.548 800.924 260.787 770.492 97
HPEIN0.618 830.729 720.668 910.647 890.597 680.766 850.414 790.680 770.520 760.525 670.946 600.432 730.215 740.493 870.599 800.638 720.617 980.570 650.897 450.806 590.605 66
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-GCNpermissive0.610 840.858 230.772 440.489 1060.532 860.792 750.404 830.643 860.570 640.507 750.935 830.414 840.046 1110.510 810.702 660.602 820.705 740.549 790.859 780.773 840.534 90
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
AttAN0.609 850.760 570.667 920.649 880.521 880.793 730.457 580.648 840.528 740.434 930.947 570.401 870.153 990.454 920.721 620.648 670.717 680.536 860.904 380.765 860.485 99
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
wsss-transformer0.600 860.634 920.743 650.697 720.601 670.781 780.437 740.585 930.493 840.446 880.933 880.394 880.011 1130.654 400.661 750.603 810.733 630.526 880.832 840.761 880.480 100
LAP-D0.594 870.720 760.692 850.637 940.456 980.773 830.391 890.730 660.587 540.445 900.940 770.381 910.288 370.434 950.453 960.591 860.649 910.581 630.777 920.749 920.610 62
DPC0.592 880.720 760.700 790.602 980.480 940.762 880.380 920.713 720.585 570.437 910.940 770.369 930.288 370.434 950.509 910.590 880.639 960.567 690.772 940.755 900.592 72
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
CCRFNet0.589 890.766 560.659 960.683 750.470 970.740 930.387 900.620 890.490 850.476 810.922 950.355 960.245 640.511 800.511 900.571 910.643 940.493 950.872 690.762 870.600 68
ROSMRF0.580 900.772 510.707 750.681 760.563 800.764 860.362 940.515 1020.465 930.465 850.936 820.427 790.207 760.438 930.577 820.536 940.675 850.486 960.723 1000.779 800.524 93
SD-DETR0.576 910.746 630.609 1050.445 1100.517 890.643 1050.366 930.714 710.456 950.468 840.870 1070.432 730.264 560.558 690.674 700.586 890.688 800.482 970.739 980.733 950.537 89
SQN_0.1%0.569 920.676 850.696 820.657 840.497 900.779 810.424 760.548 980.515 780.376 980.902 1040.422 810.357 80.379 1000.456 950.596 850.659 890.544 810.685 1030.665 1060.556 85
TextureNetpermissive0.566 930.672 870.664 930.671 800.494 920.719 950.445 670.678 790.411 1030.396 960.935 830.356 950.225 690.412 970.535 860.565 920.636 970.464 990.794 910.680 1030.568 79
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
DVVNet0.562 940.648 890.700 790.770 530.586 730.687 990.333 980.650 830.514 790.475 820.906 1010.359 940.223 710.340 1020.442 970.422 1050.668 870.501 920.708 1010.779 800.534 90
Pointnet++ & Featurepermissive0.557 950.735 680.661 950.686 740.491 930.744 920.392 870.539 990.451 960.375 990.946 600.376 920.205 780.403 980.356 1020.553 930.643 940.497 930.824 870.756 890.515 94
GMLPs0.538 960.495 1070.693 840.647 890.471 960.793 730.300 1010.477 1030.505 810.358 1010.903 1030.327 990.081 1080.472 900.529 880.448 1030.710 690.509 890.746 960.737 940.554 86
PanopticFusion-label0.529 970.491 1080.688 880.604 970.386 1030.632 1060.225 1110.705 750.434 1000.293 1070.815 1090.348 970.241 650.499 850.669 710.507 960.649 910.442 1050.796 900.602 1100.561 82
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_weak0.516 980.676 850.591 1080.609 950.442 990.774 820.335 970.597 900.422 1020.357 1020.932 890.341 980.094 1070.298 1040.528 890.473 1010.676 840.495 940.602 1090.721 980.349 110
Online SegFusion0.515 990.607 970.644 990.579 1000.434 1000.630 1070.353 950.628 880.440 980.410 940.762 1130.307 1010.167 940.520 780.403 1000.516 950.565 1010.447 1030.678 1040.701 1000.514 95
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 1000.558 1020.608 1060.424 1120.478 950.690 980.246 1070.586 920.468 910.450 870.911 990.394 880.160 970.438 930.212 1090.432 1040.541 1070.475 980.742 970.727 960.477 101
PCNN0.498 1010.559 1010.644 990.560 1020.420 1020.711 970.229 1090.414 1040.436 990.352 1030.941 750.324 1000.155 980.238 1090.387 1010.493 970.529 1080.509 890.813 890.751 910.504 96
Weakly-Openseg v30.489 1020.749 620.664 930.646 910.496 910.559 1110.122 1140.577 940.257 1140.364 1000.805 1100.198 1120.096 1060.510 810.496 930.361 1090.563 1020.359 1120.777 920.644 1070.532 92
3DMV0.484 1030.484 1090.538 1100.643 920.424 1010.606 1100.310 990.574 950.433 1010.378 970.796 1110.301 1020.214 750.537 760.208 1100.472 1020.507 1110.413 1080.693 1020.602 1100.539 88
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PointCNN with RGBpermissive0.458 1040.577 1000.611 1040.356 1140.321 1110.715 960.299 1030.376 1080.328 1100.319 1050.944 690.285 1040.164 950.216 1120.229 1070.484 990.545 1060.456 1010.755 950.709 990.475 102
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
FCPNpermissive0.447 1050.679 840.604 1070.578 1010.380 1040.682 1000.291 1040.106 1140.483 880.258 1120.920 960.258 1080.025 1120.231 1110.325 1030.480 1000.560 1040.463 1000.725 990.666 1050.231 114
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
DGCNN_reproducecopyleft0.446 1060.474 1100.623 1020.463 1080.366 1060.651 1030.310 990.389 1070.349 1080.330 1040.937 800.271 1060.126 1030.285 1050.224 1080.350 1110.577 1000.445 1040.625 1070.723 970.394 106
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 1070.548 1040.548 1090.597 990.363 1070.628 1080.300 1010.292 1090.374 1050.307 1060.881 1060.268 1070.186 870.238 1090.204 1110.407 1060.506 1120.449 1020.667 1050.620 1090.462 104
SurfaceConvPF0.442 1070.505 1060.622 1030.380 1130.342 1090.654 1020.227 1100.397 1060.367 1060.276 1090.924 930.240 1090.198 830.359 1010.262 1050.366 1070.581 990.435 1060.640 1060.668 1040.398 105
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
Tangent Convolutionspermissive0.438 1090.437 1120.646 980.474 1070.369 1050.645 1040.353 950.258 1110.282 1120.279 1080.918 980.298 1030.147 1020.283 1060.294 1040.487 980.562 1030.427 1070.619 1080.633 1080.352 109
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
3DWSSS0.425 1100.525 1050.647 970.522 1030.324 1100.488 1140.077 1150.712 730.353 1070.401 950.636 1150.281 1050.176 900.340 1020.565 840.175 1150.551 1050.398 1090.370 1150.602 1100.361 108
SPLAT Netcopyleft0.393 1110.472 1110.511 1110.606 960.311 1120.656 1010.245 1080.405 1050.328 1100.197 1130.927 920.227 1110.000 1160.001 1160.249 1060.271 1140.510 1090.383 1110.593 1100.699 1010.267 112
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 1120.297 1140.491 1120.432 1110.358 1080.612 1090.274 1050.116 1130.411 1030.265 1100.904 1020.229 1100.079 1090.250 1070.185 1120.320 1120.510 1090.385 1100.548 1110.597 1130.394 106
PointNet++permissive0.339 1130.584 990.478 1130.458 1090.256 1140.360 1150.250 1060.247 1120.278 1130.261 1110.677 1140.183 1130.117 1040.212 1130.145 1140.364 1080.346 1150.232 1150.548 1110.523 1140.252 113
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 1140.353 1130.290 1150.278 1150.166 1150.553 1120.169 1130.286 1100.147 1150.148 1150.908 1000.182 1140.064 1100.023 1150.018 1160.354 1100.363 1130.345 1130.546 1130.685 1020.278 111
ScanNetpermissive0.306 1150.203 1150.366 1140.501 1040.311 1120.524 1130.211 1120.002 1160.342 1090.189 1140.786 1120.145 1150.102 1050.245 1080.152 1130.318 1130.348 1140.300 1140.460 1140.437 1150.182 115
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 1160.000 1160.041 1160.172 1160.030 1160.062 1160.001 1160.035 1150.004 1160.051 1160.143 1160.019 1160.003 1150.041 1140.050 1150.003 1160.054 1160.018 1160.005 1160.264 1160.082 116


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




Method Infoavg ap 25%bathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
OneFormer3Dcopyleft0.896 11.000 11.000 10.913 50.858 50.951 50.786 110.837 160.916 110.908 10.778 50.803 40.750 121.000 10.976 30.926 40.882 50.995 430.849 1
Maxim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: OneFormer3D: One Transformer for Unified Point Cloud Segmentation.
MG-Former0.887 21.000 10.991 110.837 220.801 190.935 140.887 20.857 90.946 30.891 60.748 130.805 30.739 141.000 10.993 20.809 530.876 121.000 10.842 2
UniPerception0.884 31.000 10.979 170.872 150.869 20.892 230.806 80.890 50.835 260.892 50.755 100.811 10.779 90.955 430.951 40.876 210.914 10.997 360.840 3
InsSSM0.883 41.000 10.996 30.800 350.865 30.960 20.808 70.852 130.940 50.899 40.785 30.810 20.700 181.000 10.912 150.851 380.895 20.997 360.827 5
TST3D0.879 51.000 10.994 60.921 40.807 180.939 110.771 120.887 60.923 90.862 130.722 180.768 110.756 111.000 10.910 250.904 60.836 220.999 350.824 7
SIM3D0.878 61.000 10.972 210.863 170.817 160.952 40.821 50.783 270.890 150.902 30.735 160.797 50.799 81.000 10.931 120.893 120.853 181.000 10.792 13
EV3D0.877 71.000 10.996 50.873 130.854 60.950 60.691 210.783 280.926 60.889 90.754 110.794 80.820 21.000 10.912 150.900 80.860 161.000 10.779 16
TD3Dpermissive0.875 81.000 10.976 200.877 110.783 250.970 10.889 10.828 170.945 40.803 180.713 200.720 200.709 161.000 10.936 100.934 30.873 131.000 10.791 14
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
Spherical Mask(CtoF)0.875 81.000 10.991 120.873 130.850 70.946 80.691 210.752 320.926 60.889 80.759 80.794 70.820 21.000 10.912 150.900 80.878 91.000 10.769 18
Queryformer0.874 101.000 10.978 190.809 330.876 10.936 130.702 180.716 370.920 100.875 120.766 60.772 100.818 51.000 10.995 10.916 50.892 31.000 10.767 19
SoftGroup++0.874 101.000 10.972 220.947 10.839 100.898 220.556 360.913 20.881 180.756 200.828 20.748 150.821 11.000 10.937 90.937 10.887 41.000 10.821 8
Mask3D0.870 121.000 10.985 140.782 420.818 150.938 120.760 130.749 330.923 80.877 110.760 70.785 90.820 21.000 10.912 150.864 320.878 90.983 490.825 6
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
ExtMask3D0.867 131.000 11.000 10.756 490.816 170.940 100.795 90.760 310.862 200.888 100.739 140.763 120.774 101.000 10.929 130.878 200.879 71.000 10.819 10
SoftGrouppermissive0.865 141.000 10.969 230.860 180.860 40.913 180.558 330.899 30.911 120.760 190.828 10.736 170.802 70.981 400.919 140.875 220.877 111.000 10.820 9
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
MAFT0.860 151.000 10.990 130.810 320.829 110.949 70.809 60.688 430.836 250.904 20.751 120.796 60.741 131.000 10.864 350.848 400.837 201.000 10.828 4
IPCA-Inst0.851 161.000 10.968 240.884 100.842 90.862 350.693 200.812 220.888 170.677 320.783 40.698 210.807 61.000 10.911 220.865 310.865 151.000 10.757 22
SPFormerpermissive0.851 161.000 10.994 70.806 340.774 270.942 90.637 250.849 140.859 220.889 70.720 190.730 180.665 231.000 10.911 220.868 300.873 141.000 10.796 12
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
Mask3D_evaluation0.843 181.000 10.955 290.847 200.795 210.932 150.750 150.780 290.891 140.818 150.737 150.633 300.703 171.000 10.902 270.870 260.820 230.941 570.805 11
ISBNetpermissive0.835 191.000 10.950 300.731 510.819 130.918 160.790 100.740 340.851 240.831 140.661 280.742 160.650 261.000 10.937 80.814 520.836 211.000 10.765 20
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
SphereSeg0.835 191.000 10.963 270.891 80.794 220.954 30.822 40.710 380.961 20.721 240.693 260.530 430.653 251.000 10.867 340.857 350.859 170.991 460.771 17
TopoSeg0.832 211.000 10.981 160.933 20.819 140.826 440.524 420.841 150.811 300.681 310.759 90.687 220.727 150.981 400.911 220.883 160.853 191.000 10.756 23
GraphCut0.832 211.000 10.922 440.724 530.798 200.902 210.701 190.856 110.859 210.715 250.706 210.748 140.640 371.000 10.934 110.862 330.880 61.000 10.729 25
PBNetpermissive0.825 231.000 10.963 260.837 240.843 80.865 300.822 30.647 460.878 190.733 220.639 350.683 230.650 261.000 10.853 360.870 270.820 241.000 10.744 24
W.Zhao, Y.Yan, C.Yang, J.Ye,X.Yang,K.Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
SSEC0.820 241.000 10.983 150.924 30.826 120.817 470.415 510.899 40.793 340.673 330.731 170.636 280.653 241.000 10.939 70.804 550.878 81.000 10.780 15
DKNet0.815 251.000 10.930 360.844 210.765 310.915 170.534 400.805 240.805 320.807 170.654 290.763 130.650 261.000 10.794 480.881 170.766 281.000 10.758 21
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
RPGN0.806 261.000 10.992 90.789 370.723 440.891 240.650 240.810 230.832 270.665 350.699 240.658 240.700 181.000 10.881 290.832 440.774 260.997 360.613 45
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
Box2Mask0.803 271.000 10.962 280.874 120.707 480.887 270.686 230.598 510.961 10.715 260.694 250.469 480.700 181.000 10.912 150.902 70.753 330.997 360.637 39
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
HAISpermissive0.803 271.000 10.994 70.820 280.759 320.855 360.554 370.882 70.827 290.615 410.676 270.638 270.646 351.000 10.912 150.797 580.767 270.994 440.726 26
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
Mask-Group0.792 291.000 10.968 250.812 290.766 300.864 310.460 450.815 210.888 160.598 450.651 320.639 260.600 430.918 460.941 50.896 110.721 401.000 10.723 27
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
CSC-Pretrained0.791 301.000 10.996 30.829 270.767 290.889 260.600 280.819 200.770 390.594 460.620 390.541 400.700 181.000 10.941 50.889 140.763 291.000 10.526 55
SSTNetpermissive0.789 311.000 10.840 580.888 90.717 450.835 400.717 170.684 440.627 540.724 230.652 310.727 190.600 431.000 10.912 150.822 470.757 321.000 10.691 33
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
GICN0.788 321.000 10.978 180.867 160.781 260.833 410.527 410.824 180.806 310.549 540.596 420.551 360.700 181.000 10.853 360.935 20.733 371.000 10.651 36
DENet0.786 331.000 10.929 370.736 500.750 380.720 600.755 140.934 10.794 330.590 470.561 480.537 410.650 261.000 10.882 280.804 560.789 251.000 10.719 28
DANCENET0.786 331.000 10.936 330.783 400.737 410.852 380.742 160.647 460.765 410.811 160.624 380.579 330.632 401.000 10.909 260.898 100.696 450.944 530.601 48
DualGroup0.782 351.000 10.927 380.811 300.772 280.853 370.631 270.805 240.773 360.613 420.611 400.610 310.650 260.835 570.881 290.879 190.750 351.000 10.675 34
PointGroup0.778 361.000 10.900 480.798 360.715 460.863 320.493 430.706 390.895 130.569 520.701 220.576 340.639 381.000 10.880 310.851 370.719 410.997 360.709 30
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]
PE0.776 371.000 10.900 490.860 180.728 430.869 280.400 520.857 100.774 350.568 530.701 230.602 320.646 350.933 450.843 390.890 130.691 490.997 360.709 29
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
AOIA0.767 381.000 10.937 320.810 310.740 400.906 190.550 380.800 260.706 460.577 510.624 370.544 390.596 480.857 490.879 330.880 180.750 340.992 450.658 35
DD-UNet+Group0.764 391.000 10.897 510.837 230.753 350.830 430.459 470.824 180.699 480.629 390.653 300.438 510.650 261.000 10.880 310.858 340.690 501.000 10.650 37
H. Liu, R. Liu, K. Yang, J. Zhang, K. Peng, R. Stiefelhagen: HIDA: Towards Holistic Indoor Understanding for the Visually Impaired via Semantic Instance Segmentation with a Wearable Solid-State LiDAR Sensor. ICCVW 2021
INS-Conv-instance0.762 401.000 10.923 410.765 450.785 240.905 200.600 280.655 450.646 530.683 300.647 330.530 420.650 261.000 10.824 410.830 450.693 480.944 530.644 38
Dyco3Dcopyleft0.761 411.000 10.935 340.893 70.752 370.863 330.600 280.588 520.742 430.641 370.633 360.546 380.550 500.857 490.789 500.853 360.762 300.987 470.699 31
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
OccuSeg+instance0.742 421.000 10.923 410.785 380.745 390.867 290.557 340.578 550.729 440.670 340.644 340.488 460.577 491.000 10.794 480.830 450.620 581.000 10.550 51
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
RWSeg0.739 431.000 10.899 500.759 470.753 360.823 450.282 570.691 420.658 510.582 500.594 430.547 370.628 411.000 10.795 470.868 290.728 391.000 10.692 32
3D-MPA0.737 441.000 10.933 350.785 380.794 230.831 420.279 590.588 520.695 490.616 400.559 490.556 350.650 261.000 10.809 450.875 230.696 461.000 10.608 47
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
MTML0.731 451.000 10.992 90.779 440.609 570.746 550.308 560.867 80.601 570.607 430.539 520.519 440.550 501.000 10.824 410.869 280.729 381.000 10.616 43
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
OSIS0.725 461.000 10.885 540.653 590.657 540.801 480.576 320.695 410.828 280.698 280.534 530.457 500.500 570.857 490.831 400.841 420.627 561.000 10.619 42
SSEN0.724 471.000 10.926 390.781 430.661 520.845 390.596 310.529 580.764 420.653 360.489 590.461 490.500 570.859 480.765 510.872 250.761 311.000 10.577 49
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
NeuralBF0.718 481.000 10.945 310.901 60.754 340.817 460.460 450.700 400.772 370.688 290.568 470.000 700.500 570.981 400.606 610.872 240.740 361.000 10.614 44
Weiwei Sun, Daniel Rebain, Renjie Liao, Vladimir Tankovich, Soroosh Yazdani, Kwang Moo Yi, Andrea Tagliasacchi: NeuralBF: Neural Bilateral Filtering for Top-down Instance Segmentation on Point Clouds. WACV 2023
Sparse R-CNN0.714 491.000 10.926 400.694 540.699 500.890 250.636 260.516 590.693 500.743 210.588 440.369 550.601 420.594 630.800 460.886 150.676 510.986 480.546 52
SALoss-ResNet0.695 501.000 10.855 560.579 640.589 590.735 580.484 440.588 520.856 230.634 380.571 460.298 560.500 571.000 10.824 410.818 480.702 440.935 600.545 53
Zhidong Liang, Ming Yang, Hao Li, Chunxiang Wang: 3D Instance Embedding Learning With a Structure-Aware Loss Function for Point Cloud Segmentation. IEEE Robotics and Automation Letters (IROS2020)
PanopticFusion-inst0.693 511.000 10.852 570.655 580.616 560.788 500.334 540.763 300.771 380.457 640.555 500.652 250.518 540.857 490.765 510.732 640.631 540.944 530.577 50
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
Occipital-SCS0.688 521.000 10.913 450.730 520.737 420.743 570.442 480.855 120.655 520.546 550.546 510.263 580.508 560.889 470.568 620.771 610.705 430.889 630.625 41
3D-BoNet0.687 531.000 10.887 530.836 250.587 600.643 670.550 380.620 480.724 450.522 590.501 570.243 590.512 551.000 10.751 530.807 540.661 530.909 620.612 46
Bo Yang, Jianan Wang, Ronald Clark, Qingyong Hu, Sen Wang, Andrew Markham, Niki Trigoni: Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds. NeurIPS 2019 Spotlight
ClickSeg_Instance0.685 541.000 10.818 600.600 620.715 470.795 490.557 340.533 570.591 590.601 440.519 550.429 530.638 390.938 440.706 560.817 500.624 570.944 530.502 57
PCJC0.684 551.000 10.895 520.757 480.659 530.862 340.189 660.739 350.606 560.712 270.581 450.515 450.650 260.857 490.357 670.785 590.631 550.889 630.635 40
SPG_WSIS0.678 561.000 10.880 550.836 250.701 490.727 590.273 610.607 500.706 470.541 570.515 560.174 620.600 430.857 490.716 550.846 410.711 421.000 10.506 56
One_Thing_One_Clickpermissive0.675 571.000 10.823 590.782 410.621 550.766 520.211 630.736 360.560 610.586 480.522 540.636 290.453 610.641 610.853 360.850 390.694 470.997 360.411 62
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
SegGroup_inspermissive0.637 581.000 10.923 430.593 630.561 610.746 560.143 680.504 600.766 400.485 620.442 600.372 540.530 530.714 580.815