Presenting the ScanNet200 Benchmark

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

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

ScanNet200 Benchmark

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




Method Infoavg iouhead ioucommon ioutail ioualarm clockarmchairbackpackbagballbarbasketbathroom cabinetbathroom counterbathroom stallbathroom stall doorbathroom vanitybathtubbedbenchbicyclebinblackboardblanketblindsboardbookbookshelfbottlebowlboxbroombucketbulletin boardcabinetcalendarcandlecartcase of water bottlescd caseceilingceiling lightchairclockclosetcloset doorcloset rodcloset wallclothesclothes dryercoat rackcoffee kettlecoffee makercoffee tablecolumncomputer towercontainercopiercouchcountercratecupcurtaincushiondecorationdeskdining tabledish rackdishwasherdividerdoordoorframedresserdumbbelldustpanend tablefanfile cabinetfire alarmfire extinguisherfireplacefloorfolded chairfurnitureguitarguitar casehair dryerhandicap barhatheadphonesironing boardjacketkeyboardkeyboard pianokitchen cabinetkitchen counterladderlamplaptoplaundry basketlaundry detergentlaundry hamperledgelightlight switchluggagemachinemailboxmatmattressmicrowavemini fridgemirrormonitormousemusic standnightstandobjectoffice chairottomanovenpaperpaper bagpaper cutterpaper towel dispenserpaper towel rollpersonpianopicturepillarpillowpipeplantplateplungerposterpotted plantpower outletpower stripprinterprojectorprojector screenpurserackradiatorrailrange hoodrecycling binrefrigeratorscaleseatshelfshoeshowershower curtainshower curtain rodshower doorshower floorshower headshower wallsignsinksoap dishsoap dispensersofa chairspeakerstair railstairsstandstoolstorage binstorage containerstorage organizerstovestructurestuffed animalsuitcasetabletelephonetissue boxtoastertoaster oventoilettoilet papertoilet paper dispensertoilet paper holdertoilet seat cover dispensertoweltrash bintrash cantraytubetvtv standvacuum cleanerventwallwardrobewashing machinewater bottlewater coolerwater pitcherwhiteboardwindowwindowsill
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort 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 by
AWCS0.305 50.508 50.225 50.142 20.463 50.063 60.195 50.000 10.000 10.000 10.467 20.551 10.504 30.773 30.764 50.142 50.029 80.000 30.626 60.100 50.000 40.360 50.179 60.507 60.137 60.006 30.300 50.000 10.000 30.172 30.364 60.512 20.000 10.056 50.000 20.865 50.093 30.634 80.000 30.071 60.396 60.296 70.876 40.000 10.000 10.373 50.436 70.063 60.749 10.877 20.721 30.131 20.124 50.804 60.000 10.000 40.515 40.010 40.452 50.252 40.578 50.417 20.179 80.484 40.171 30.337 50.606 50.000 20.115 30.937 50.142 40.000 10.008 30.000 70.157 70.484 50.402 80.501 60.339 30.553 10.529 20.478 50.000 20.000 10.404 30.001 40.022 60.077 20.000 30.894 60.219 10.628 20.093 60.305 50.886 10.233 20.000 10.603 30.112 30.023 50.000 40.000 10.000 30.741 20.664 20.097 60.253 50.782 50.264 30.523 40.154 10.707 70.000 30.411 30.000 10.000 20.000 20.332 70.000 10.000 40.000 10.602 20.595 40.185 70.656 70.159 10.000 10.355 50.424 60.154 60.729 60.516 50.220 40.620 20.084 50.000 10.707 50.651 40.173 10.014 50.381 80.582 70.000 10.619 20.049 60.000 40.000 10.702 10.000 20.000 10.302 70.489 60.317 40.334 40.392 10.922 50.254 50.533 60.394 40.129 80.613 60.000 10.000 60.820 10.649 50.749 50.000 10.782 50.282 30.863 30.000 10.288 70.006 50.220 50.633 50.542 1
OA-CNN-L_ScanNet2000.333 20.558 10.269 20.124 40.448 60.080 40.272 30.000 10.000 10.000 10.342 50.515 20.524 20.713 80.789 20.158 40.384 30.000 30.806 20.125 20.000 40.496 30.332 20.498 70.227 40.024 20.474 10.000 10.003 20.071 40.487 10.000 30.000 10.110 20.000 20.876 10.013 80.703 10.000 30.076 40.473 40.355 40.906 20.000 10.000 10.476 40.706 10.000 70.672 60.835 50.748 20.015 70.223 20.860 30.000 10.000 40.572 20.000 50.509 30.313 20.662 10.398 50.396 10.411 60.276 10.527 10.711 10.000 20.076 50.946 10.166 20.000 10.022 20.160 10.183 40.493 40.699 30.637 20.403 20.330 50.406 40.526 20.024 10.000 10.392 40.000 50.016 80.000 30.196 20.915 20.112 30.557 30.197 10.352 40.877 20.000 30.000 10.592 60.103 60.000 80.067 10.000 10.089 10.735 30.625 30.130 50.568 20.836 20.271 10.534 30.043 60.799 20.001 20.445 10.000 10.000 20.024 10.661 20.000 10.262 10.000 10.591 30.517 70.373 30.788 30.021 20.000 10.455 10.517 40.320 30.823 40.200 80.001 80.150 30.100 40.000 10.736 20.668 20.103 60.052 30.662 10.720 10.000 10.602 40.112 30.002 30.000 10.637 40.000 20.000 10.621 30.569 10.398 20.412 30.234 30.949 10.363 10.492 70.495 30.251 30.665 30.000 10.001 50.805 20.833 20.794 40.000 10.821 10.314 20.843 50.000 10.560 20.245 20.262 20.713 10.370 5
PPT-SpUNet-F.T.0.332 30.556 20.270 10.123 50.519 10.091 20.349 20.000 10.000 10.000 10.339 60.383 50.498 50.833 20.807 10.241 10.584 20.000 30.755 30.124 30.000 40.608 20.330 30.530 40.314 10.000 40.374 30.000 10.000 30.197 10.459 30.000 30.000 10.117 10.000 20.876 10.095 10.682 20.000 30.086 30.518 20.433 10.930 10.000 10.000 10.563 30.542 50.077 40.715 20.858 30.756 10.008 80.171 40.874 20.000 10.039 10.550 30.000 50.545 20.256 30.657 30.453 10.351 30.449 50.213 20.392 30.611 40.000 20.037 60.946 10.138 50.000 10.000 40.063 30.308 10.537 20.796 10.673 10.323 50.392 30.400 50.509 30.000 20.000 10.649 10.000 50.023 50.000 30.000 30.914 30.002 70.506 70.163 50.359 30.872 30.000 30.000 10.623 20.112 30.001 70.000 40.000 10.021 20.753 10.565 70.150 10.579 10.806 40.267 20.616 10.042 70.783 40.000 30.374 50.000 10.000 20.000 20.620 40.000 10.000 40.000 10.572 60.634 10.350 40.792 20.000 30.000 10.376 40.535 20.378 20.855 10.672 10.074 50.000 40.185 30.000 10.727 30.660 30.076 80.000 60.432 40.646 30.000 10.594 50.006 70.000 40.000 10.658 20.000 20.000 10.661 10.549 30.300 50.291 50.045 50.942 40.304 20.600 30.572 20.135 70.695 10.000 10.008 30.793 30.942 10.899 20.000 10.816 20.181 40.897 10.000 10.679 10.223 30.264 10.691 20.345 6
Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao: Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training.
OctFormer ScanNet200permissive0.326 40.539 40.265 30.131 30.499 20.110 10.522 10.000 10.000 10.000 10.318 80.427 40.455 60.743 60.765 40.175 30.842 10.000 30.828 10.204 10.033 20.429 40.335 10.601 10.312 20.000 40.357 40.000 10.000 30.047 50.423 40.000 30.000 10.105 30.000 20.873 30.079 60.670 40.000 30.117 10.471 50.432 20.829 50.000 10.000 10.584 20.417 80.089 30.684 50.837 40.705 70.021 60.178 30.892 10.000 10.028 20.505 50.000 50.457 40.200 60.662 10.412 30.244 60.496 30.000 80.451 20.626 30.000 20.102 40.943 30.138 50.000 10.000 40.149 20.291 20.534 30.722 20.632 30.331 40.253 70.453 30.487 40.000 20.000 10.479 20.000 50.022 60.000 30.000 30.900 40.128 20.684 10.164 40.413 10.854 50.000 30.000 10.512 80.074 80.003 60.000 40.000 10.000 30.469 60.613 40.132 40.529 30.871 10.227 70.582 20.026 80.787 30.000 30.339 60.000 10.000 20.000 20.626 30.000 10.029 30.000 10.587 40.612 30.411 20.724 50.000 30.000 10.407 30.552 10.513 10.849 20.655 20.408 10.000 40.296 10.000 10.686 60.645 50.145 30.022 40.414 50.633 40.000 10.637 10.224 10.000 40.000 10.650 30.000 20.000 10.622 20.535 40.343 30.483 20.230 40.943 30.289 30.618 20.596 10.140 60.679 20.000 10.022 10.783 40.620 60.906 10.000 10.806 30.137 60.865 20.000 10.378 40.000 70.168 80.680 30.227 7
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
LGroundpermissive0.272 60.485 60.184 60.106 60.476 30.077 50.218 40.000 10.000 10.000 10.547 10.295 60.540 10.746 50.745 60.058 70.112 70.005 10.658 50.077 80.000 40.322 60.178 70.512 50.190 50.199 10.277 60.000 10.000 30.173 20.399 50.000 30.000 10.039 70.000 20.858 60.085 50.676 30.002 10.103 20.498 30.323 50.703 60.000 10.000 10.296 60.549 40.216 10.702 30.768 60.718 50.028 40.092 70.786 70.000 10.000 40.453 70.022 30.251 80.252 40.572 60.348 60.321 40.514 20.063 60.279 70.552 60.000 20.019 70.932 60.132 70.000 10.000 40.000 70.156 80.457 60.623 50.518 50.265 70.358 40.381 60.395 60.000 20.000 10.127 80.012 30.051 10.000 30.000 30.886 70.014 50.437 80.179 20.244 60.826 60.000 30.000 10.599 40.136 10.085 20.000 40.000 10.000 30.565 40.612 50.143 20.207 60.566 60.232 60.446 60.127 20.708 60.000 30.384 40.000 10.000 20.000 20.402 50.000 10.059 20.000 10.525 80.566 50.229 60.659 60.000 30.000 10.265 60.446 50.147 70.720 80.597 40.066 60.000 40.187 20.000 10.726 40.467 80.134 50.000 60.413 60.629 50.000 10.363 70.055 50.022 20.000 10.626 50.000 20.000 10.323 60.479 80.154 70.117 60.028 70.901 60.243 60.415 80.295 80.143 50.610 70.000 10.000 60.777 50.397 80.324 70.000 10.778 60.179 50.702 70.000 10.274 80.404 10.233 40.622 60.398 3
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
CeCo0.340 10.551 30.247 40.181 10.475 40.057 80.142 70.000 10.000 10.000 10.387 30.463 30.499 40.924 10.774 30.213 20.257 40.000 30.546 80.100 50.006 30.615 10.177 80.534 20.246 30.000 40.400 20.000 10.338 10.006 70.484 20.609 10.000 10.083 40.000 20.873 30.089 40.661 50.000 30.048 80.560 10.408 30.892 30.000 10.000 10.586 10.616 30.000 70.692 40.900 10.721 30.162 10.228 10.860 30.000 10.000 40.575 10.083 20.550 10.347 10.624 40.410 40.360 20.740 10.109 50.321 60.660 20.000 20.121 20.939 40.143 30.000 10.400 10.003 50.190 30.564 10.652 40.615 40.421 10.304 60.579 10.547 10.000 20.000 10.296 50.000 50.030 40.096 10.000 30.916 10.037 40.551 40.171 30.376 20.865 40.286 10.000 10.633 10.102 70.027 40.011 30.000 10.000 30.474 50.742 10.133 30.311 40.824 30.242 40.503 50.068 40.828 10.000 30.429 20.000 10.063 10.000 20.781 10.000 10.000 40.000 10.665 10.633 20.450 10.818 10.000 30.000 10.429 20.532 30.226 40.825 30.510 60.377 20.709 10.079 60.000 10.753 10.683 10.102 70.063 20.401 70.620 60.000 10.619 20.000 80.000 40.000 10.595 60.000 20.000 10.345 50.564 20.411 10.603 10.384 20.945 20.266 40.643 10.367 50.304 10.663 40.000 10.010 20.726 60.767 30.898 30.000 10.784 40.435 10.861 40.000 10.447 30.000 70.257 30.656 40.377 4
: Understanding Imbalanced Semantic Segmentation Through Neural Collapse.
Minkowski 34Dpermissive0.253 70.463 70.154 80.102 70.381 80.084 30.134 80.000 10.000 10.000 10.386 40.141 80.279 80.737 70.703 70.014 80.164 60.000 30.663 40.092 70.000 40.224 70.291 40.531 30.056 80.000 40.242 70.000 10.000 30.013 60.331 70.000 30.000 10.035 80.001 10.858 60.059 70.650 70.000 30.056 70.353 70.299 60.670 70.000 10.000 10.284 70.484 60.071 50.594 70.720 70.710 60.027 50.068 80.813 50.000 10.005 30.492 60.164 10.274 70.111 80.571 70.307 80.293 50.307 80.150 40.163 80.531 70.002 10.545 10.932 60.093 80.000 10.000 40.002 60.159 60.368 80.581 70.440 80.228 80.406 20.282 80.294 70.000 20.000 10.189 70.060 10.036 30.000 30.000 30.897 50.000 80.525 60.025 80.205 80.771 80.000 30.000 10.593 50.108 50.044 30.000 40.000 10.000 30.282 80.589 60.094 70.169 70.466 80.227 70.419 80.125 30.757 50.002 10.334 70.000 10.000 20.000 20.357 60.000 10.000 40.000 10.582 50.513 80.337 50.612 80.000 30.000 10.250 70.352 80.136 80.724 70.655 20.280 30.000 40.046 80.000 10.606 80.559 60.159 20.102 10.445 20.655 20.000 10.310 80.117 20.000 40.000 10.581 80.026 10.000 10.265 80.483 70.084 80.097 80.044 60.865 80.142 80.588 40.351 60.272 20.596 80.000 10.003 40.622 70.720 40.096 80.000 10.771 70.016 70.772 60.000 10.302 60.194 40.214 60.621 70.197 8
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
CSC-Pretrainpermissive0.249 80.455 80.171 70.079 80.418 70.059 70.186 60.000 10.000 10.000 10.335 70.250 70.316 70.766 40.697 80.142 50.170 50.003 20.553 70.112 40.097 10.201 80.186 50.476 80.081 70.000 40.216 80.000 10.000 30.001 80.314 80.000 30.000 10.055 60.000 20.832 80.094 20.659 60.002 10.076 40.310 80.293 80.664 80.000 10.000 10.175 80.634 20.130 20.552 80.686 80.700 80.076 30.110 60.770 80.000 10.000 40.430 80.000 50.319 60.166 70.542 80.327 70.205 70.332 70.052 70.375 40.444 80.000 20.012 80.930 80.203 10.000 10.000 40.046 40.175 50.413 70.592 60.471 70.299 60.152 80.340 70.247 80.000 20.000 10.225 60.058 20.037 20.000 30.207 10.862 80.014 50.548 50.033 70.233 70.816 70.000 30.000 10.542 70.123 20.121 10.019 20.000 10.000 30.463 70.454 80.045 80.128 80.557 70.235 50.441 70.063 50.484 80.000 30.308 80.000 10.000 20.000 20.318 80.000 10.000 40.000 10.545 70.543 60.164 80.734 40.000 30.000 10.215 80.371 70.198 50.743 50.205 70.062 70.000 40.079 60.000 10.683 70.547 70.142 40.000 60.441 30.579 80.000 10.464 60.098 40.041 10.000 10.590 70.000 20.000 10.373 40.494 50.174 60.105 70.001 80.895 70.222 70.537 50.307 70.180 40.625 50.000 10.000 60.591 80.609 70.398 60.000 10.766 80.014 80.638 80.000 10.377 50.004 60.206 70.609 80.465 2
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021


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




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


ScanNet Benchmark

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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort bysort by
OccuSeg+Semantic0.764 50.758 520.796 230.839 150.746 170.907 10.562 70.850 180.680 110.672 90.978 20.610 20.335 110.777 50.819 370.847 10.830 10.691 100.972 10.885 40.727 15
PPT-SpUNet-Joint0.766 30.932 20.794 250.829 190.751 150.854 110.540 140.903 40.630 260.672 90.963 90.565 150.357 40.788 20.900 80.737 190.802 130.685 120.950 20.887 20.780 2
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.
IPCA0.731 240.890 100.837 30.864 20.726 240.873 30.530 190.824 290.489 780.647 130.978 20.609 30.336 100.624 420.733 510.758 130.776 310.570 570.949 30.877 80.728 13
PointConvFormer0.749 110.793 350.790 280.807 300.750 160.856 100.524 200.881 90.588 460.642 190.977 40.591 60.274 370.781 30.929 20.804 30.796 190.642 260.947 40.885 40.715 22
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
One-Thing-One-Click0.693 360.743 580.794 250.655 780.684 340.822 410.497 340.719 590.622 280.617 280.977 40.447 600.339 90.750 210.664 670.703 340.790 260.596 470.946 50.855 240.647 42
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
PointTransformerV20.752 90.742 590.809 150.872 10.758 90.860 80.552 90.891 80.610 340.687 30.960 110.559 180.304 230.766 100.926 30.767 100.797 180.644 250.942 60.876 110.722 19
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
Retro-FPN0.744 170.842 220.800 200.767 470.740 190.836 270.541 130.914 10.672 130.626 250.958 140.552 210.272 390.777 50.886 150.696 360.801 160.674 170.941 70.858 200.717 20
Peng Xiang*, Xin Wen*, Yu-Shen Liu, Hui Zhang, Yi Fang, Zhizhong Han: Retrospective Feature Pyramid Network for Point Cloud Semantic Segmentation. ICCV 2023
Mix3Dpermissive0.781 10.964 10.855 10.843 130.781 30.858 90.575 40.831 250.685 90.714 20.979 10.594 40.310 200.801 10.892 120.841 20.819 30.723 30.940 80.887 20.725 17
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
VMNetpermissive0.746 150.870 140.838 20.858 40.729 230.850 160.501 290.874 110.587 470.658 120.956 190.564 160.299 240.765 110.900 80.716 280.812 90.631 310.939 90.858 200.709 23
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)
MatchingNet0.724 270.812 320.812 130.810 280.735 210.834 290.495 350.860 160.572 530.602 360.954 280.512 330.280 340.757 150.845 310.725 220.780 290.606 430.937 100.851 290.700 28
OA-CNN-L_ScanNet200.756 80.783 380.826 40.858 40.776 40.837 250.548 110.896 70.649 190.675 70.962 100.586 90.335 110.771 80.802 410.770 90.787 270.691 100.936 110.880 70.761 6
Virtual MVFusion0.746 150.771 460.819 80.848 90.702 290.865 70.397 760.899 50.699 40.664 110.948 470.588 70.330 130.746 220.851 290.764 110.796 190.704 80.935 120.866 150.728 13
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
SparseConvNet0.725 250.647 820.821 60.846 100.721 250.869 40.533 160.754 490.603 400.614 290.955 220.572 130.325 150.710 270.870 170.724 230.823 20.628 320.934 130.865 160.683 32
INS-Conv-semantic0.717 290.751 550.759 440.812 260.704 280.868 50.537 150.842 220.609 360.608 320.953 310.534 260.293 270.616 450.864 200.719 270.793 230.640 270.933 140.845 340.663 37
LargeKernel3D0.739 210.909 60.820 70.806 320.740 190.852 140.545 120.826 270.594 450.643 160.955 220.541 230.263 480.723 260.858 240.775 80.767 370.678 140.933 140.848 300.694 29
Yukang Chen*, Jianhui Liu*, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs. CVPR 2023
O-CNNpermissive0.762 70.924 40.823 50.844 120.770 50.852 140.577 20.847 210.711 10.640 200.958 140.592 50.217 640.762 130.888 130.758 130.813 80.726 10.932 160.868 130.744 8
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
PointTransformer++0.725 250.727 670.811 140.819 220.765 70.841 220.502 280.814 340.621 290.623 270.955 220.556 190.284 320.620 440.866 190.781 60.757 460.648 230.932 160.862 170.709 23
CU-Hybrid Net0.764 50.924 40.819 80.840 140.757 100.853 130.580 10.848 190.709 20.643 160.958 140.587 80.295 260.753 180.884 160.758 130.815 50.725 20.927 180.867 140.743 9
RPN0.736 220.776 420.790 280.851 60.754 120.854 110.491 380.866 130.596 440.686 40.955 220.536 240.342 80.624 420.869 180.787 50.802 130.628 320.927 180.875 120.704 26
OctFormerpermissive0.766 30.925 30.808 160.849 70.786 20.846 190.566 60.876 100.690 70.674 80.960 110.576 110.226 580.753 180.904 60.777 70.815 50.722 40.923 200.877 80.776 4
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
SAT0.742 190.860 170.765 410.819 220.769 60.848 170.533 160.829 260.663 150.631 230.955 220.586 90.274 370.753 180.896 100.729 200.760 430.666 200.921 210.855 240.733 11
MSP0.748 130.623 850.804 180.859 30.745 180.824 380.501 290.912 20.690 70.685 50.956 190.567 140.320 170.768 90.918 40.720 250.802 130.676 150.921 210.881 60.779 3
DMF-Net0.752 90.906 80.793 270.802 340.689 320.825 360.556 80.867 120.681 100.602 360.960 110.555 200.365 30.779 40.859 220.747 160.795 220.717 50.917 230.856 220.764 5
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
Swin3Dpermissive0.779 20.861 160.818 100.836 160.790 10.875 20.576 30.905 30.704 30.739 10.969 60.611 10.349 60.756 160.958 10.702 350.805 120.708 60.916 240.898 10.801 1
BPNetcopyleft0.749 110.909 60.818 100.811 270.752 130.839 240.485 390.842 220.673 120.644 150.957 170.528 290.305 220.773 70.859 220.788 40.818 40.693 90.916 240.856 220.723 18
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
LRPNet0.742 190.816 300.806 170.807 300.752 130.828 340.575 40.839 240.699 40.637 210.954 280.520 310.320 170.755 170.834 330.760 120.772 330.676 150.915 260.862 170.717 20
PointContrast_LA_SEM0.683 440.757 530.784 320.786 390.639 480.824 380.408 710.775 430.604 390.541 510.934 790.532 270.269 430.552 630.777 440.645 610.793 230.640 270.913 270.824 410.671 35
dtc_net0.596 770.683 750.725 620.715 590.549 750.803 590.444 600.647 750.493 740.495 680.941 660.409 750.000 1050.424 870.544 750.598 740.703 670.522 790.912 280.792 640.520 84
Feature-Geometry Netpermissive0.685 400.866 150.748 500.819 220.645 460.794 640.450 530.802 380.587 470.604 340.945 550.464 500.201 720.554 620.840 320.723 240.732 550.602 450.907 290.822 440.603 58
EQ-Net0.743 180.620 860.799 220.849 70.730 220.822 410.493 360.897 60.664 140.681 60.955 220.562 170.378 10.760 140.903 70.738 180.801 160.673 180.907 290.877 80.745 7
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
Feature_GeometricNetpermissive0.690 380.884 120.754 480.795 370.647 440.818 480.422 680.802 380.612 330.604 340.945 550.462 510.189 770.563 590.853 270.726 210.765 380.632 300.904 310.821 450.606 55
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
AttAN0.609 750.760 510.667 830.649 800.521 790.793 650.457 490.648 740.528 650.434 840.947 490.401 780.153 900.454 820.721 550.648 570.717 590.536 760.904 310.765 770.485 89
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
PicassoNet-IIpermissive0.692 370.732 630.772 370.786 390.677 360.866 60.517 230.848 190.509 700.626 250.952 350.536 240.225 600.545 660.704 580.689 410.810 100.564 610.903 330.854 270.729 12
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
One Thing One Click0.701 340.825 280.796 230.723 540.716 260.832 300.433 660.816 310.634 240.609 310.969 60.418 730.344 70.559 600.833 340.715 290.808 110.560 620.902 340.847 310.680 33
PointConvpermissive0.666 490.781 390.759 440.699 630.644 470.822 410.475 400.779 420.564 580.504 670.953 310.428 670.203 710.586 510.754 470.661 510.753 470.588 530.902 340.813 490.642 43
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointNet2-SFPN0.631 660.771 460.692 760.672 710.524 780.837 250.440 630.706 650.538 620.446 790.944 610.421 720.219 630.552 630.751 490.591 770.737 520.543 730.901 360.768 760.557 75
StratifiedFormerpermissive0.747 140.901 90.803 190.845 110.757 100.846 190.512 240.825 280.696 60.645 140.956 190.576 110.262 490.744 230.861 210.742 170.770 360.705 70.899 370.860 190.734 10
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
HPEIN0.618 730.729 650.668 820.647 810.597 590.766 770.414 700.680 680.520 660.525 580.946 520.432 630.215 650.493 780.599 720.638 620.617 890.570 570.897 380.806 510.605 57
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
FusionAwareConv0.630 690.604 890.741 570.766 480.590 610.747 830.501 290.734 560.503 720.527 570.919 890.454 550.323 160.550 650.420 880.678 430.688 720.544 710.896 390.795 580.627 49
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
Superpoint Network0.683 440.851 200.728 610.800 360.653 420.806 560.468 440.804 360.572 530.602 360.946 520.453 570.239 570.519 710.822 350.689 410.762 420.595 490.895 400.827 400.630 48
PointMRNet0.640 580.717 700.701 690.692 660.576 670.801 600.467 460.716 600.563 590.459 770.953 310.429 660.169 840.581 520.854 260.605 700.710 600.550 690.894 410.793 610.575 66
SALANet0.670 480.816 300.770 390.768 460.652 430.807 550.451 500.747 510.659 180.545 500.924 850.473 470.149 920.571 560.811 390.635 640.746 500.623 350.892 420.794 590.570 68
PNE0.721 280.840 230.789 300.833 170.690 300.823 400.509 250.864 150.618 300.629 240.957 170.500 360.266 460.763 120.797 430.674 480.791 250.621 370.892 420.855 240.708 25
RFCR0.702 330.889 110.745 530.813 250.672 370.818 480.493 360.815 330.623 270.610 300.947 490.470 480.249 530.594 480.848 300.705 330.779 300.646 240.892 420.823 420.611 51
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
contrastBoundarypermissive0.705 310.769 490.775 360.809 290.687 330.820 440.439 640.812 350.661 160.591 420.945 550.515 320.171 820.633 390.856 250.720 250.796 190.668 190.889 450.847 310.689 30
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
VI-PointConv0.676 460.770 480.754 480.783 420.621 520.814 510.552 90.758 470.571 550.557 470.954 280.529 280.268 450.530 690.682 620.675 440.719 580.603 440.888 460.833 360.665 36
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
SConv0.636 620.830 260.697 720.752 520.572 690.780 720.445 570.716 600.529 640.530 560.951 370.446 610.170 830.507 750.666 660.636 630.682 740.541 740.886 470.799 540.594 62
ROSMRF3D0.673 470.789 360.748 500.763 490.635 500.814 510.407 730.747 510.581 510.573 440.950 410.484 420.271 410.607 460.754 470.649 550.774 320.596 470.883 480.823 420.606 55
KP-FCNN0.684 410.847 210.758 460.784 410.647 440.814 510.473 410.772 440.605 380.594 410.935 750.450 580.181 800.587 490.805 400.690 390.785 280.614 390.882 490.819 460.632 47
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
VACNN++0.684 410.728 660.757 470.776 440.690 300.804 580.464 470.816 310.577 520.587 430.945 550.508 350.276 360.671 290.710 560.663 500.750 490.589 520.881 500.832 380.653 40
DenSeR0.628 700.800 330.625 910.719 570.545 760.806 560.445 570.597 810.448 880.519 620.938 710.481 430.328 140.489 790.499 830.657 530.759 440.592 500.881 500.797 570.634 46
ClickSeg_Semantic0.703 320.774 440.800 200.793 380.760 80.847 180.471 420.802 380.463 850.634 220.968 80.491 400.271 410.726 250.910 50.706 320.815 50.551 680.878 520.833 360.570 68
MVPNetpermissive0.641 560.831 250.715 630.671 730.590 610.781 700.394 770.679 690.642 200.553 480.937 720.462 510.256 500.649 330.406 890.626 650.691 710.666 200.877 530.792 640.608 54
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
SIConv0.625 720.830 260.694 740.757 500.563 710.772 760.448 540.647 750.520 660.509 630.949 450.431 650.191 760.496 770.614 710.647 580.672 780.535 770.876 540.783 700.571 67
PointConv-SFPN0.641 560.776 420.703 670.721 560.557 730.826 350.451 500.672 710.563 590.483 710.943 630.425 700.162 870.644 350.726 520.659 520.709 620.572 560.875 550.786 690.559 74
3DSM_DMMF0.631 660.626 840.745 530.801 350.607 550.751 820.506 260.729 580.565 570.491 700.866 990.434 620.197 750.595 470.630 690.709 310.705 650.560 620.875 550.740 840.491 88
MinkowskiNetpermissive0.736 220.859 180.818 100.832 180.709 270.840 230.521 220.853 170.660 170.643 160.951 370.544 220.286 310.731 240.893 110.675 440.772 330.683 130.874 570.852 280.727 15
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
APCF-Net0.631 660.742 590.687 810.672 710.557 730.792 670.408 710.665 720.545 610.508 640.952 350.428 670.186 780.634 380.702 590.620 660.706 640.555 660.873 580.798 560.581 64
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
SegGroup_sempermissive0.627 710.818 290.747 520.701 620.602 570.764 780.385 820.629 780.490 760.508 640.931 820.409 750.201 720.564 580.725 530.618 670.692 700.539 750.873 580.794 590.548 78
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
CCRFNet0.589 800.766 500.659 860.683 680.470 870.740 850.387 810.620 800.490 760.476 730.922 870.355 870.245 550.511 720.511 810.571 820.643 850.493 860.872 600.762 780.600 59
FPConvpermissive0.639 590.785 370.760 430.713 610.603 560.798 620.392 780.534 910.603 400.524 590.948 470.457 530.250 520.538 670.723 540.598 740.696 690.614 390.872 600.799 540.567 71
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
JSENetpermissive0.699 350.881 130.762 420.821 200.667 380.800 610.522 210.792 410.613 320.607 330.935 750.492 390.205 690.576 530.853 270.691 380.758 450.652 220.872 600.828 390.649 41
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
DGNet0.684 410.712 710.784 320.782 430.658 390.835 280.499 330.823 300.641 210.597 390.950 410.487 410.281 330.575 540.619 700.647 580.764 390.620 380.871 630.846 330.688 31
PPCNN++permissive0.663 510.746 560.708 650.722 550.638 490.820 440.451 500.566 860.599 420.541 510.950 410.510 340.313 190.648 340.819 370.616 690.682 740.590 510.869 640.810 500.656 39
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
PointMTL0.632 650.731 640.688 790.675 700.591 600.784 690.444 600.565 870.610 340.492 690.949 450.456 540.254 510.587 490.706 570.599 730.665 800.612 420.868 650.791 680.579 65
PD-Net0.638 600.797 340.769 400.641 840.590 610.820 440.461 480.537 900.637 230.536 540.947 490.388 810.206 680.656 310.668 650.647 580.732 550.585 540.868 650.793 610.473 93
Supervoxel-CNN0.635 630.656 800.711 640.719 570.613 540.757 810.444 600.765 460.534 630.566 450.928 830.478 450.272 390.636 360.531 780.664 490.645 840.508 820.864 670.792 640.611 51
FusionNet0.688 390.704 720.741 570.754 510.656 400.829 320.501 290.741 540.609 360.548 490.950 410.522 300.371 20.633 390.756 460.715 290.771 350.623 350.861 680.814 470.658 38
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
SPH3D-GCNpermissive0.610 740.858 190.772 370.489 960.532 770.792 670.404 740.643 770.570 560.507 660.935 750.414 740.046 1010.510 730.702 590.602 720.705 650.549 700.859 690.773 750.534 81
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
SAFNet-segpermissive0.654 540.752 540.734 590.664 760.583 650.815 500.399 750.754 490.639 220.535 550.942 640.470 480.309 210.665 300.539 760.650 540.708 630.635 290.857 700.793 610.642 43
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
HPGCNN0.656 530.698 740.743 550.650 790.564 700.820 440.505 270.758 470.631 250.479 720.945 550.480 440.226 580.572 550.774 450.690 390.735 530.614 390.853 710.776 740.597 61
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
PointMetaBase0.714 300.835 240.785 310.821 200.684 340.846 190.531 180.865 140.614 310.596 400.953 310.500 360.246 540.674 280.888 130.692 370.764 390.624 340.849 720.844 350.675 34
PointSPNet0.637 610.734 620.692 760.714 600.576 670.797 630.446 550.743 530.598 430.437 820.942 640.403 770.150 910.626 410.800 420.649 550.697 680.557 650.846 730.777 730.563 72
joint point-basedpermissive0.634 640.614 870.778 350.667 750.633 510.825 360.420 690.804 360.467 830.561 460.951 370.494 380.291 280.566 570.458 840.579 810.764 390.559 640.838 740.814 470.598 60
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
wsss-transformer0.600 760.634 830.743 550.697 650.601 580.781 700.437 650.585 840.493 740.446 790.933 800.394 790.011 1030.654 320.661 680.603 710.733 540.526 780.832 750.761 790.480 90
RandLA-Netpermissive0.645 550.778 400.731 600.699 630.577 660.829 320.446 550.736 550.477 800.523 610.945 550.454 550.269 430.484 800.749 500.618 670.738 510.599 460.827 760.792 640.621 50
DCM-Net0.658 520.778 400.702 680.806 320.619 530.813 540.468 440.693 670.494 730.524 590.941 660.449 590.298 250.510 730.821 360.675 440.727 570.568 590.826 770.803 530.637 45
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
Pointnet++ & Featurepermissive0.557 860.735 610.661 850.686 670.491 830.744 840.392 780.539 890.451 870.375 900.946 520.376 830.205 690.403 890.356 920.553 840.643 850.497 840.824 780.756 800.515 85
PointASNLpermissive0.666 490.703 730.781 340.751 530.655 410.830 310.471 420.769 450.474 810.537 530.951 370.475 460.279 350.635 370.698 610.675 440.751 480.553 670.816 790.806 510.703 27
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PCNN0.498 920.559 920.644 890.560 930.420 920.711 880.229 1000.414 940.436 900.352 930.941 660.324 910.155 890.238 990.387 910.493 880.529 980.509 800.813 800.751 820.504 87
PanopticFusion-label0.529 880.491 980.688 790.604 880.386 930.632 970.225 1020.705 660.434 910.293 970.815 1000.348 880.241 560.499 760.669 640.507 870.649 820.442 960.796 810.602 1000.561 73
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
TextureNetpermissive0.566 840.672 790.664 840.671 730.494 820.719 860.445 570.678 700.411 940.396 870.935 750.356 860.225 600.412 880.535 770.565 830.636 880.464 900.794 820.680 940.568 70
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
LAP-D0.594 780.720 680.692 760.637 850.456 880.773 750.391 800.730 570.587 470.445 810.940 690.381 820.288 290.434 850.453 860.591 770.649 820.581 550.777 830.749 830.610 53
DPC0.592 790.720 680.700 700.602 890.480 840.762 800.380 830.713 630.585 500.437 820.940 690.369 840.288 290.434 850.509 820.590 790.639 870.567 600.772 840.755 810.592 63
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
PointCNN with RGBpermissive0.458 940.577 910.611 940.356 1040.321 1010.715 870.299 940.376 980.328 1010.319 950.944 610.285 950.164 860.216 1020.229 970.484 900.545 960.456 920.755 850.709 900.475 92
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
GMLPs0.538 870.495 970.693 750.647 810.471 860.793 650.300 920.477 930.505 710.358 910.903 950.327 900.081 980.472 810.529 790.448 940.710 600.509 800.746 860.737 850.554 77
3DMV, FTSDF0.501 910.558 930.608 960.424 1020.478 850.690 890.246 980.586 830.468 820.450 780.911 910.394 790.160 880.438 830.212 990.432 950.541 970.475 890.742 870.727 870.477 91
SD-DETR0.576 820.746 560.609 950.445 1000.517 800.643 960.366 840.714 620.456 860.468 750.870 980.432 630.264 470.558 610.674 630.586 800.688 720.482 880.739 880.733 860.537 80
FCPNpermissive0.447 950.679 760.604 970.578 920.380 940.682 910.291 950.106 1040.483 790.258 1020.920 880.258 990.025 1020.231 1010.325 930.480 910.560 940.463 910.725 890.666 960.231 104
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
ROSMRF0.580 810.772 450.707 660.681 690.563 710.764 780.362 850.515 920.465 840.465 760.936 740.427 690.207 670.438 830.577 730.536 850.675 770.486 870.723 900.779 710.524 83
DVVNet0.562 850.648 810.700 700.770 450.586 640.687 900.333 890.650 730.514 690.475 740.906 930.359 850.223 620.340 920.442 870.422 960.668 790.501 830.708 910.779 710.534 81
3DMV0.484 930.484 990.538 1000.643 830.424 910.606 1010.310 900.574 850.433 920.378 880.796 1010.301 930.214 660.537 680.208 1000.472 930.507 1010.413 990.693 920.602 1000.539 79
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
SQN_0.1%0.569 830.676 770.696 730.657 770.497 810.779 730.424 670.548 880.515 680.376 890.902 960.422 710.357 40.379 900.456 850.596 760.659 810.544 710.685 930.665 970.556 76
Online SegFusion0.515 900.607 880.644 890.579 910.434 900.630 980.353 860.628 790.440 890.410 850.762 1030.307 920.167 850.520 700.403 900.516 860.565 920.447 940.678 940.701 910.514 86
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
PNET20.442 970.548 940.548 990.597 900.363 970.628 990.300 920.292 990.374 960.307 960.881 970.268 980.186 780.238 990.204 1010.407 970.506 1020.449 930.667 950.620 990.462 94
SurfaceConvPF0.442 970.505 960.622 930.380 1030.342 990.654 930.227 1010.397 960.367 970.276 990.924 850.240 1000.198 740.359 910.262 950.366 980.581 900.435 970.640 960.668 950.398 95
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
DGCNN_reproducecopyleft0.446 960.474 1000.623 920.463 980.366 960.651 940.310 900.389 970.349 990.330 940.937 720.271 970.126 940.285 950.224 980.350 1010.577 910.445 950.625 970.723 880.394 96
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
Tangent Convolutionspermissive0.438 990.437 1020.646 880.474 970.369 950.645 950.353 860.258 1010.282 1030.279 980.918 900.298 940.147 930.283 960.294 940.487 890.562 930.427 980.619 980.633 980.352 99
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
subcloud_weak0.516 890.676 770.591 980.609 860.442 890.774 740.335 880.597 810.422 930.357 920.932 810.341 890.094 970.298 940.528 800.473 920.676 760.495 850.602 990.721 890.349 100
SPLAT Netcopyleft0.393 1010.472 1010.511 1010.606 870.311 1020.656 920.245 990.405 950.328 1010.197 1030.927 840.227 1020.000 1050.001 1060.249 960.271 1040.510 990.383 1020.593 1000.699 920.267 102
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 1020.297 1040.491 1020.432 1010.358 980.612 1000.274 960.116 1030.411 940.265 1000.904 940.229 1010.079 990.250 970.185 1020.320 1020.510 990.385 1010.548 1010.597 1030.394 96
PointNet++permissive0.339 1030.584 900.478 1030.458 990.256 1040.360 1050.250 970.247 1020.278 1040.261 1010.677 1040.183 1030.117 950.212 1030.145 1040.364 990.346 1050.232 1050.548 1010.523 1040.252 103
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 1040.353 1030.290 1050.278 1050.166 1050.553 1020.169 1040.286 1000.147 1050.148 1050.908 920.182 1040.064 1000.023 1050.018 1060.354 1000.363 1030.345 1030.546 1030.685 930.278 101
ScanNetpermissive0.306 1050.203 1050.366 1040.501 950.311 1020.524 1030.211 1030.002 1060.342 1000.189 1040.786 1020.145 1050.102 960.245 980.152 1030.318 1030.348 1040.300 1040.460 1040.437 1050.182 105
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
3DWSSS0.425 1000.525 950.647 870.522 940.324 1000.488 1040.077 1050.712 640.353 980.401 860.636 1050.281 960.176 810.340 920.565 740.175 1050.551 950.398 1000.370 1050.602 1000.361 98
ERROR0.054 1060.000 1060.041 1060.172 1060.030 1060.062 1060.001 1060.035 1050.004 1060.051 1060.143 1060.019 1060.003 1040.041 1040.050 1050.003 1060.054 1060.018 1060.005 1060.264 1060.082 106


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




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort by
ISBNetpermissive0.559 50.926 20.597 70.390 230.436 40.722 80.276 30.556 160.380 40.450 60.505 80.583 30.730 10.575 280.455 60.603 160.573 40.979 10.332 14
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
PBNetpermissive0.573 30.926 20.575 110.619 10.472 20.736 50.239 60.487 260.383 30.459 40.506 70.533 80.585 70.767 80.404 90.717 30.559 50.969 20.381 5
W.Zhao, Y.Yan, C.Yang, J.Ye,X.Yang,K.Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
Queryformer0.583 20.926 20.702 20.393 220.504 10.733 70.276 20.527 200.373 50.479 20.534 40.533 90.697 30.720 160.436 80.745 20.592 10.958 30.363 8
SSTNetpermissive0.506 110.738 180.549 160.497 50.316 200.693 130.178 80.377 340.198 230.330 130.463 140.576 50.515 120.857 30.494 10.637 130.457 160.943 40.290 24
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
MAFT0.596 10.889 60.721 10.448 110.460 30.768 10.251 40.558 150.408 10.504 10.539 20.616 10.618 50.858 20.482 30.684 70.551 60.931 50.450 1
Mask3D0.566 40.926 20.597 60.408 190.420 50.737 40.239 50.598 80.386 20.458 50.549 10.568 60.716 20.601 270.480 40.646 100.575 30.922 60.364 7
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
TopoSeg0.479 150.704 210.564 120.467 90.366 130.633 270.068 180.554 170.262 170.328 140.447 160.323 140.534 110.722 150.288 210.614 140.482 140.912 70.358 11
TD3D0.489 130.852 80.511 240.434 150.322 190.735 60.101 140.512 220.355 60.349 120.468 130.283 200.514 130.676 220.268 240.671 80.510 90.908 80.329 16
SoftGrouppermissive0.504 120.667 280.579 80.372 260.381 90.694 120.072 170.677 20.303 110.387 90.531 50.319 160.582 80.754 100.318 150.643 110.492 130.907 90.388 3
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
GraphCut0.552 61.000 10.611 50.438 140.392 80.714 90.139 90.598 90.327 80.389 80.510 60.598 20.427 240.754 110.463 50.761 10.588 20.903 100.329 15
DKNet0.532 80.815 100.624 40.517 30.377 100.749 20.107 110.509 230.304 100.437 70.475 100.581 40.539 100.775 70.339 140.640 120.506 100.901 110.385 4
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
SSEC0.465 170.667 280.578 90.502 40.362 160.641 260.035 300.605 60.291 160.323 160.451 150.296 180.417 270.677 210.245 280.501 360.506 110.900 120.366 6
DualGroup0.469 160.815 100.552 140.398 200.374 110.683 150.130 100.539 190.310 90.327 150.407 190.276 210.447 190.535 320.342 130.659 90.455 170.900 130.301 20
SPFormerpermissive0.549 70.745 150.640 30.484 60.395 70.739 30.311 10.566 130.335 70.468 30.492 90.555 70.478 150.747 130.436 70.712 40.540 70.893 140.343 13
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
DENet0.413 240.741 160.520 200.237 390.284 230.523 380.097 150.691 10.138 280.209 380.229 400.238 270.390 290.707 180.310 160.448 430.470 150.892 150.310 18
HAISpermissive0.457 180.704 210.561 130.457 100.364 140.673 160.046 290.547 180.194 240.308 170.426 170.288 190.454 180.711 170.262 250.563 260.434 200.889 160.344 12
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
Box2Mask0.433 220.741 160.463 340.433 160.283 240.625 290.103 120.298 440.125 330.260 230.424 180.322 150.472 160.701 190.363 120.711 50.309 420.882 170.272 28
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
CSC-Pretrained0.405 260.738 180.465 330.331 320.205 370.655 220.051 250.601 70.092 380.211 370.329 280.198 320.459 170.775 60.195 350.524 340.400 250.878 180.184 39
IPCA-Inst0.520 90.889 60.551 150.548 20.418 60.665 190.064 200.585 100.260 180.277 210.471 120.500 100.644 40.785 50.369 100.591 190.511 80.878 190.362 9
Mask-Group0.434 210.778 130.516 220.471 80.330 180.658 200.029 320.526 210.249 190.256 240.400 200.309 170.384 310.296 480.368 110.575 220.425 210.877 200.362 10
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
SoftGroup++0.513 100.704 210.578 100.398 210.363 150.704 100.061 210.647 40.297 150.378 110.537 30.343 120.614 60.828 40.295 190.710 60.505 120.875 210.394 2
RWSeg0.348 370.475 440.456 350.320 330.275 270.476 400.020 380.491 250.056 450.212 350.320 290.261 230.302 360.520 330.182 370.557 280.285 440.867 220.197 36
OSIS0.392 290.778 130.530 190.220 410.278 250.567 350.083 160.330 400.299 130.270 220.310 310.143 380.260 380.624 250.277 220.568 250.361 300.865 230.301 19
PointGroup0.407 250.639 350.496 280.415 180.243 330.645 250.021 370.570 120.114 340.211 360.359 250.217 310.428 230.660 230.256 260.562 270.341 340.860 240.291 22
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]
AOIA0.387 300.704 210.515 230.385 240.225 360.669 170.005 450.482 270.126 320.181 410.269 370.221 300.426 250.478 380.218 310.592 180.371 280.851 250.242 32
Dyco3Dcopyleft0.395 280.642 340.518 210.447 120.259 300.666 180.050 260.251 480.166 260.231 290.362 240.232 280.331 340.535 310.229 290.587 200.438 190.850 260.317 17
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
3D-MPA0.355 350.457 470.484 310.299 340.277 260.591 340.047 280.332 370.212 210.217 320.278 330.193 340.413 280.410 420.195 340.574 240.352 310.849 270.213 35
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
DD-UNet+Group0.436 190.630 360.508 270.480 70.310 210.624 300.065 190.638 50.174 250.256 250.384 220.194 330.428 220.759 90.289 200.574 230.400 240.849 280.291 23
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
One_Thing_One_Clickpermissive0.326 390.472 450.361 400.232 400.183 400.555 360.000 510.498 240.038 470.195 390.226 410.362 110.168 460.469 400.251 270.553 290.335 360.846 290.117 47
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
OccuSeg+instance0.486 140.802 120.536 180.428 170.369 120.702 110.205 70.331 390.301 120.379 100.474 110.327 130.437 200.862 10.485 20.601 170.394 260.846 300.273 26
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
SSEN0.384 310.852 80.494 300.192 420.226 350.648 240.022 350.398 320.299 140.277 200.317 300.231 290.194 450.514 350.196 330.586 210.444 180.843 310.184 38
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
RPGN0.428 230.630 360.508 260.367 270.249 310.658 210.016 390.673 30.131 310.234 280.383 230.270 220.434 210.748 120.274 230.609 150.406 230.842 320.267 29
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
INS-Conv-instance0.435 200.716 200.495 290.355 280.331 170.689 140.102 130.394 330.208 220.280 190.395 210.250 240.544 90.741 140.309 170.536 320.391 270.842 330.258 30
ClickSeg_Instance0.366 330.654 320.375 380.184 430.302 220.592 330.050 270.300 430.093 370.283 180.277 340.249 250.426 260.615 260.299 180.504 350.367 290.832 340.191 37
PE0.396 270.667 280.467 320.446 130.243 320.624 310.022 360.577 110.106 350.219 310.340 260.239 260.487 140.475 390.225 300.541 310.350 320.818 350.273 27
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
NeuralBF0.353 360.593 380.511 250.375 250.264 280.597 320.008 410.332 380.160 270.229 300.274 360.000 580.206 420.678 200.155 420.485 380.422 220.816 360.254 31
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
Occipital-SCS0.320 400.679 270.352 410.334 310.229 340.436 420.025 330.412 310.058 430.161 460.240 390.085 420.262 370.496 370.187 360.467 400.328 380.775 370.231 34
SPG_WSIS0.251 460.380 490.274 480.289 350.144 430.413 450.000 510.311 410.065 410.113 480.130 480.029 500.204 430.388 430.108 480.459 410.311 400.769 380.127 46
Sparse R-CNN0.292 410.704 210.213 510.153 450.154 420.551 370.053 230.212 490.132 300.174 430.274 350.070 440.363 320.441 410.176 380.424 450.234 460.758 390.161 43
SegGroup_inspermissive0.246 470.556 410.335 430.062 540.115 470.490 390.000 510.297 450.018 510.186 400.142 460.083 430.233 390.216 500.153 430.469 390.251 450.744 400.083 50
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
PCJC0.375 320.704 210.542 170.284 360.197 390.649 230.006 430.426 280.138 290.242 260.304 320.183 360.388 300.629 240.141 450.546 300.344 330.738 410.283 25
GICN0.341 380.580 390.371 390.344 300.198 380.469 410.052 240.564 140.093 360.212 340.212 420.127 400.347 330.537 300.206 320.525 330.329 370.729 420.241 33
SphereSeg0.357 340.651 330.411 360.345 290.264 290.630 280.059 220.289 460.212 200.240 270.336 270.158 370.305 350.557 290.159 410.455 420.341 350.726 430.294 21
SALoss-ResNet0.262 430.667 280.335 420.067 520.123 460.427 440.022 340.280 470.058 420.216 330.211 430.039 480.142 480.519 340.106 490.338 490.310 410.721 440.138 44
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)
MASCpermissive0.254 440.463 460.249 500.113 460.167 410.412 460.000 500.374 350.073 390.173 440.243 380.130 390.228 410.368 440.160 400.356 470.208 470.711 450.136 45
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
PanopticFusion-inst0.214 480.250 530.330 440.275 370.103 490.228 570.000 510.345 360.024 490.088 500.203 440.186 350.167 470.367 450.125 460.221 550.112 570.666 460.162 42
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
3D-BoNet0.253 450.519 420.324 450.251 380.137 450.345 510.031 310.419 300.069 400.162 450.131 470.052 460.202 440.338 460.147 440.301 520.303 430.651 470.178 40
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
MTML0.282 420.577 400.380 370.182 440.107 480.430 430.001 480.422 290.057 440.179 420.162 450.070 450.229 400.511 360.161 390.491 370.313 390.650 480.162 41
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
UNet-backbone0.161 490.519 420.259 490.084 480.059 510.325 530.002 460.093 540.009 530.077 520.064 510.045 470.044 550.161 520.045 510.331 500.180 490.566 490.033 58
R-PointNet0.158 510.356 500.173 530.113 470.140 440.359 470.012 400.023 560.039 460.134 470.123 490.008 540.089 510.149 530.117 470.221 540.128 540.563 500.094 48
Region-18class0.146 520.175 570.321 460.080 490.062 500.357 480.000 510.307 420.002 550.066 530.044 530.000 580.018 570.036 570.054 500.447 440.133 520.472 510.060 53
3D-SISpermissive0.161 490.407 480.155 550.068 510.043 550.346 500.001 470.134 510.005 540.088 490.106 500.037 490.135 500.321 470.028 540.339 480.116 560.466 520.093 49
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
SemRegionNet-20cls0.121 530.296 520.203 520.071 500.058 520.349 490.000 510.150 500.019 500.054 540.034 550.017 530.052 530.042 560.013 570.209 560.183 480.371 530.057 54
3D-BEVIS0.117 540.250 530.308 470.020 580.009 590.269 560.006 440.008 570.029 480.037 570.014 580.003 560.036 560.147 540.042 520.381 460.118 550.362 540.069 52
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.
tmp0.113 560.333 510.151 560.056 550.053 530.344 520.000 510.105 530.016 520.049 550.035 540.020 520.053 520.048 550.013 560.183 570.173 500.344 550.054 55
Sgpn_scannet0.049 580.023 590.134 570.031 570.013 580.144 580.006 420.008 580.000 580.028 580.017 570.003 550.009 590.000 580.021 550.122 580.095 580.175 560.054 56
ASIS0.085 570.037 580.080 580.066 530.047 540.282 550.000 510.052 550.002 560.047 560.026 560.001 570.046 540.194 510.031 530.264 530.140 510.167 570.047 57
Hier3Dcopyleft0.117 540.222 550.161 540.054 560.027 560.289 540.000 510.124 520.001 570.079 510.061 520.027 510.141 490.240 490.005 580.310 510.129 530.153 580.081 51
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
MaskRCNN 2d->3d Proj0.022 590.185 560.000 590.000 590.015 570.000 590.000 490.006 590.000 580.010 590.006 590.107 410.012 580.000 580.002 590.027 590.004 590.022 590.001 59


This table lists the benchmark results for the 2D semantic label scenario.


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort bysort by
Virtual MVFusion (R)0.745 10.861 10.839 10.881 10.672 10.512 10.422 150.898 10.723 10.714 10.954 20.454 10.509 10.773 10.895 10.756 10.820 10.653 10.935 10.891 10.728 1
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
SSMAcopyleft0.577 110.695 40.716 120.439 140.563 110.314 110.444 130.719 80.551 100.503 90.887 120.346 130.348 90.603 90.353 170.709 50.600 120.457 120.901 20.786 80.599 11
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019
FAN_NV_RVC0.586 80.510 180.764 40.079 230.620 70.330 80.494 70.753 40.573 80.556 40.884 130.405 30.303 140.718 20.452 110.672 120.658 50.509 40.898 30.813 60.727 2
CMX0.613 40.681 70.725 90.502 120.634 50.297 150.478 90.830 20.651 40.537 60.924 40.375 50.315 120.686 50.451 120.714 40.543 180.504 50.894 40.823 40.688 3
MIX6D_RVC0.582 100.695 40.687 140.225 180.632 60.328 100.550 10.748 50.623 50.494 130.890 110.350 120.254 200.688 40.454 100.716 30.597 140.489 80.881 50.768 130.575 12
BPNet_2Dcopyleft0.670 20.822 30.795 30.836 20.659 20.481 20.451 110.769 30.656 30.567 30.931 30.395 40.390 40.700 30.534 30.689 90.770 20.574 30.865 60.831 30.675 4
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia and Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
DCRedNet0.583 90.682 60.723 100.542 110.510 170.310 120.451 110.668 130.549 110.520 80.920 60.375 50.446 20.528 170.417 130.670 140.577 150.478 100.862 70.806 70.628 8
UNIV_CNP_RVC_UE0.566 120.569 160.686 160.435 150.524 140.294 160.421 160.712 90.543 120.463 150.872 140.320 140.363 70.611 80.477 90.686 100.627 90.443 150.862 70.775 110.639 5
segfomer with 6d0.542 160.594 120.687 140.146 210.579 100.308 130.515 50.703 100.472 180.498 110.868 150.369 70.282 150.589 120.390 140.701 80.556 170.416 180.860 90.759 150.539 16
SN_RN152pyrx8_RVCcopyleft0.546 140.572 140.663 180.638 70.518 150.298 140.366 210.633 180.510 150.446 170.864 160.296 170.267 170.542 160.346 180.704 70.575 160.431 160.853 100.766 140.630 7
EMSAFormer0.564 130.581 130.736 80.564 100.546 130.219 200.517 40.675 110.486 170.427 190.904 90.352 110.320 110.589 120.528 40.708 60.464 210.413 190.847 110.786 80.611 10
MCA-Net0.595 60.533 170.756 60.746 40.590 80.334 70.506 60.670 120.587 70.500 100.905 80.366 80.352 80.601 100.506 60.669 150.648 70.501 60.839 120.769 120.516 18
CU-Hybrid-2D Net0.636 30.825 20.820 20.179 200.648 30.463 30.549 20.742 60.676 20.628 20.961 10.420 20.379 50.684 60.381 150.732 20.723 30.599 20.827 130.851 20.634 6
UDSSEG_RVC0.545 150.610 110.661 190.588 80.556 120.268 180.482 80.642 170.572 90.475 140.836 200.312 150.367 60.630 70.189 200.639 170.495 200.452 130.826 140.756 170.541 14
ILC-PSPNet0.475 210.490 200.581 210.289 170.507 180.067 230.379 190.610 200.417 210.435 180.822 220.278 180.267 170.503 190.228 190.616 200.533 190.375 200.820 150.729 180.560 13
RFBNet0.592 70.616 90.758 50.659 50.581 90.330 80.469 100.655 150.543 120.524 70.924 40.355 100.336 100.572 140.479 80.671 130.648 70.480 90.814 160.814 50.614 9
AdapNet++copyleft0.503 180.613 100.722 110.418 160.358 230.337 60.370 200.479 210.443 190.368 210.907 70.207 200.213 220.464 210.525 50.618 190.657 60.450 140.788 170.721 200.408 22
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. International Journal of Computer Vision, 2019
DMMF_3d0.605 50.651 80.744 70.782 30.637 40.387 40.536 30.732 70.590 60.540 50.856 180.359 90.306 130.596 110.539 20.627 180.706 40.497 70.785 180.757 160.476 19
FuseNetpermissive0.535 170.570 150.681 170.182 190.512 160.290 170.431 140.659 140.504 160.495 120.903 100.308 160.428 30.523 180.365 160.676 110.621 110.470 110.762 190.779 100.541 14
Caner Hazirbas, Lingni Ma, Csaba Domokos, Daniel Cremers: FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-based CNN Architecture. ACCV 2016
3DMV (2d proj)0.498 190.481 210.612 200.579 90.456 190.343 50.384 180.623 190.525 140.381 200.845 190.254 190.264 190.557 150.182 210.581 210.598 130.429 170.760 200.661 220.446 21
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
MSeg1080_RVCpermissive0.485 200.505 190.709 130.092 220.427 200.241 190.411 170.654 160.385 230.457 160.861 170.053 230.279 160.503 190.481 70.645 160.626 100.365 210.748 210.725 190.529 17
John Lambert*, Zhuang Liu*, Ozan Sener, James Hays, Vladlen Koltun: MSeg: A Composite Dataset for Multi-domain Semantic Segmentation. CVPR 2020
Enet (reimpl)0.376 220.264 230.452 230.452 130.365 210.181 210.143 230.456 220.409 220.346 220.769 230.164 210.218 210.359 220.123 230.403 230.381 230.313 230.571 220.685 210.472 20
Re-implementation of Adam Paszke, Abhishek Chaurasia, Sangpil Kim, Eugenio Culurciello: ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation.
ScanNet (2d proj)permissive0.330 230.293 220.521 220.657 60.361 220.161 220.250 220.004 230.440 200.183 230.836 200.125 220.060 230.319 230.132 220.417 220.412 220.344 220.541 230.427 230.109 23
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
DMMF0.003 240.000 240.005 240.000 240.000 240.037 240.001 240.000 240.001 240.005 240.003 240.000 240.000 240.000 240.000 240.000 240.002 240.001 240.000 240.006 240.000 24


This table lists the benchmark results for the 2D semantic instance scenario.




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort by
UniDet_RVC0.205 10.381 10.323 10.037 10.226 10.177 10.063 10.277 10.120 10.067 10.131 10.074 20.317 10.080 10.235 10.289 10.141 10.678 10.080 1
MaskRCNN_ScanNetpermissive0.119 20.129 20.212 20.002 20.112 20.148 20.014 20.205 20.044 20.066 20.078 20.095 10.142 20.030 20.128 20.139 20.080 20.459 20.057 2
Re-implementation of Kaiming He, Georgia Gkioxari, Piotr Dollár, Ross Girshick: Mask R-CNN. ICCV'17


This table lists the benchmark results for the scene type classification scenario.




Method Infoavg recallapartmentbathroombedroom / hotelbookstore / libraryconference roomcopy/mail roomhallwaykitchenlaundry roomliving room / loungemiscofficestorage / basement / garage
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
multi-taskpermissive0.700 10.500 11.000 10.882 20.500 21.000 11.000 10.500 21.000 11.000 10.778 10.000 20.938 10.000 2
Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler: Indoor Scene Recognition in 3D. IROS 2020
3DASPP-SCE0.691 20.500 10.938 20.824 31.000 11.000 10.500 21.000 10.857 20.500 20.556 30.000 20.812 20.500 1
SE-ResNeXt-SSMA0.498 30.000 40.812 30.941 10.500 20.500 30.500 20.500 20.429 40.500 20.667 20.500 10.625 30.000 2
Abhinav Valada, Rohit Mohan, Wolfram Burgard: Self-Supervised Model Adaptation for Multimodal Semantic Segmentation. arXiv
resnet50_scannet0.353 40.250 30.812 30.529 40.500 20.500 30.000 40.500 20.571 30.000 40.556 30.000 20.375 40.000 2