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 bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
CSC-Pretrainpermissive0.249 50.455 50.171 40.079 50.418 40.059 40.186 30.000 10.000 10.000 10.335 50.250 40.316 40.766 30.697 50.142 20.170 20.003 20.553 40.112 10.097 10.201 50.186 20.476 50.081 40.000 30.216 50.000 10.000 20.001 50.314 50.000 30.000 10.055 30.000 20.832 50.094 10.659 30.002 10.076 20.310 50.293 50.664 50.000 10.000 10.175 50.634 10.130 20.552 50.686 50.700 50.076 30.110 30.770 50.000 10.000 20.430 50.000 50.319 30.166 40.542 50.327 40.205 40.332 40.052 50.375 10.444 50.000 20.012 50.930 50.203 10.000 10.000 30.046 10.175 20.413 40.592 30.471 40.299 30.152 50.340 40.247 50.000 10.000 10.225 30.058 20.037 20.000 30.207 10.862 50.014 30.548 30.033 40.233 40.816 40.000 30.000 10.542 50.123 20.121 10.019 10.000 10.000 10.463 40.454 50.045 50.128 50.557 40.235 30.441 40.063 50.484 50.000 20.308 50.000 10.000 20.000 10.318 50.000 10.000 20.000 10.545 40.543 40.164 50.734 20.000 20.000 10.215 50.371 40.198 20.743 20.205 50.062 50.000 30.079 30.000 10.683 40.547 40.142 30.000 40.441 20.579 50.000 10.464 30.098 20.041 10.000 10.590 40.000 20.000 10.373 10.494 20.174 30.105 40.001 50.895 40.222 40.537 30.307 40.180 30.625 20.000 10.000 30.591 50.609 40.398 30.000 10.766 50.014 50.638 50.000 10.377 20.004 40.206 50.609 50.465 2
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
LGroundpermissive0.272 30.485 30.184 30.106 30.476 10.077 20.218 10.000 10.000 10.000 10.547 10.295 30.540 10.746 40.745 30.058 40.112 40.005 10.658 20.077 50.000 30.322 30.178 40.512 30.190 20.199 10.277 30.000 10.000 20.173 10.399 20.000 30.000 10.039 40.000 20.858 30.085 40.676 10.002 10.103 10.498 20.323 20.703 30.000 10.000 10.296 30.549 30.216 10.702 20.768 30.718 30.028 40.092 40.786 40.000 10.000 20.453 40.022 30.251 50.252 20.572 30.348 30.321 20.514 20.063 40.279 40.552 30.000 20.019 40.932 30.132 40.000 10.000 30.000 40.156 50.457 30.623 20.518 20.265 40.358 30.381 30.395 30.000 10.000 10.127 50.012 30.051 10.000 30.000 20.886 40.014 30.437 50.179 10.244 30.826 30.000 30.000 10.599 30.136 10.085 20.000 30.000 10.000 10.565 20.612 30.143 10.207 30.566 30.232 40.446 30.127 20.708 30.000 20.384 30.000 10.000 20.000 10.402 20.000 10.059 10.000 10.525 50.566 30.229 30.659 30.000 20.000 10.265 30.446 20.147 40.720 50.597 20.066 40.000 30.187 10.000 10.726 20.467 50.134 40.000 40.413 30.629 20.000 10.363 40.055 30.022 20.000 10.626 20.000 20.000 10.323 30.479 50.154 40.117 30.028 40.901 30.243 30.415 50.295 50.143 40.610 40.000 10.000 30.777 20.397 50.324 40.000 10.778 30.179 30.702 40.000 10.274 50.404 10.233 20.622 30.398 3
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
Minkowski 34Dpermissive0.253 40.463 40.154 50.102 40.381 50.084 10.134 50.000 10.000 10.000 10.386 40.141 50.279 50.737 50.703 40.014 50.164 30.000 30.663 10.092 40.000 30.224 40.291 10.531 20.056 50.000 30.242 40.000 10.000 20.013 30.331 40.000 30.000 10.035 50.001 10.858 30.059 50.650 40.000 30.056 40.353 40.299 30.670 40.000 10.000 10.284 40.484 40.071 30.594 40.720 40.710 40.027 50.068 50.813 20.000 10.005 10.492 30.164 10.274 40.111 50.571 40.307 50.293 30.307 50.150 20.163 50.531 40.002 10.545 10.932 30.093 50.000 10.000 30.002 30.159 30.368 50.581 40.440 50.228 50.406 20.282 50.294 40.000 10.000 10.189 40.060 10.036 30.000 30.000 20.897 20.000 50.525 40.025 50.205 50.771 50.000 30.000 10.593 40.108 40.044 30.000 30.000 10.000 10.282 50.589 40.094 40.169 40.466 50.227 50.419 50.125 30.757 20.002 10.334 40.000 10.000 20.000 10.357 30.000 10.000 20.000 10.582 30.513 50.337 20.612 50.000 20.000 10.250 40.352 50.136 50.724 40.655 10.280 20.000 30.046 50.000 10.606 50.559 30.159 20.102 10.445 10.655 10.000 10.310 50.117 10.000 30.000 10.581 50.026 10.000 10.265 50.483 40.084 50.097 50.044 30.865 50.142 50.588 20.351 30.272 20.596 50.000 10.003 20.622 40.720 20.096 50.000 10.771 40.016 40.772 30.000 10.302 30.194 20.214 40.621 40.197 5
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
CeCo0.340 10.551 10.247 10.181 10.475 20.057 50.142 40.000 10.000 10.000 10.387 30.463 20.499 30.924 10.774 10.213 10.257 10.000 30.546 50.100 20.006 20.615 10.177 50.534 10.246 10.000 30.400 10.000 10.338 10.006 40.484 10.609 10.000 10.083 10.000 20.873 10.089 30.661 20.000 30.048 50.560 10.408 10.892 10.000 10.000 10.586 10.616 20.000 50.692 30.900 10.721 10.162 10.228 10.860 10.000 10.000 20.575 10.083 20.550 10.347 10.624 10.410 20.360 10.740 10.109 30.321 30.660 10.000 20.121 20.939 10.143 20.000 10.400 10.003 20.190 10.564 10.652 10.615 10.421 10.304 40.579 10.547 10.000 10.000 10.296 20.000 50.030 40.096 10.000 20.916 10.037 20.551 20.171 20.376 10.865 20.286 10.000 10.633 10.102 50.027 40.011 20.000 10.000 10.474 30.742 10.133 20.311 10.824 10.242 20.503 20.068 40.828 10.000 20.429 10.000 10.063 10.000 10.781 10.000 10.000 20.000 10.665 10.633 10.450 10.818 10.000 20.000 10.429 10.532 10.226 10.825 10.510 40.377 10.709 10.079 30.000 10.753 10.683 10.102 50.063 20.401 40.620 30.000 10.619 10.000 50.000 30.000 10.595 30.000 20.000 10.345 20.564 10.411 10.603 10.384 20.945 10.266 10.643 10.367 20.304 10.663 10.000 10.010 10.726 30.767 10.898 10.000 10.784 10.435 10.861 20.000 10.447 10.000 50.257 10.656 10.377 4
: Understanding Imbalanced Semantic Segmentation Through Neural Collapse.
AWCS0.305 20.508 20.225 20.142 20.463 30.063 30.195 20.000 10.000 10.000 10.467 20.551 10.504 20.773 20.764 20.142 20.029 50.000 30.626 30.100 20.000 30.360 20.179 30.507 40.137 30.006 20.300 20.000 10.000 20.172 20.364 30.512 20.000 10.056 20.000 20.865 20.093 20.634 50.000 30.071 30.396 30.296 40.876 20.000 10.000 10.373 20.436 50.063 40.749 10.877 20.721 10.131 20.124 20.804 30.000 10.000 20.515 20.010 40.452 20.252 20.578 20.417 10.179 50.484 30.171 10.337 20.606 20.000 20.115 30.937 20.142 30.000 10.008 20.000 40.157 40.484 20.402 50.501 30.339 20.553 10.529 20.478 20.000 10.000 10.404 10.001 40.022 50.077 20.000 20.894 30.219 10.628 10.093 30.305 20.886 10.233 20.000 10.603 20.112 30.023 50.000 30.000 10.000 10.741 10.664 20.097 30.253 20.782 20.264 10.523 10.154 10.707 40.000 20.411 20.000 10.000 20.000 10.332 40.000 10.000 20.000 10.602 20.595 20.185 40.656 40.159 10.000 10.355 20.424 30.154 30.729 30.516 30.220 30.620 20.084 20.000 10.707 30.651 20.173 10.014 30.381 50.582 40.000 10.619 10.049 40.000 30.000 10.702 10.000 20.000 10.302 40.489 30.317 20.334 20.392 10.922 20.254 20.533 40.394 10.129 50.613 30.000 10.000 30.820 10.649 30.749 20.000 10.782 20.282 20.863 10.000 10.288 40.006 30.220 30.633 20.542 1


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 bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
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
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
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
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
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.


ScanNet Benchmark

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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
PointTransformerV20.752 50.742 530.809 140.872 10.758 50.860 60.552 70.891 50.610 300.687 20.960 80.559 140.304 200.766 80.926 20.767 80.797 140.644 220.942 60.876 70.722 14
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
IPCA0.731 200.890 80.837 30.864 20.726 180.873 20.530 150.824 230.489 730.647 90.978 20.609 20.336 70.624 370.733 460.758 110.776 250.570 530.949 20.877 50.728 8
MSP0.748 90.623 790.804 160.859 30.745 120.824 330.501 230.912 20.690 60.685 30.956 140.567 110.320 130.768 70.918 30.720 220.802 100.676 120.921 180.881 40.779 1
VMNetpermissive0.746 110.870 120.838 20.858 40.729 170.850 120.501 230.874 80.587 420.658 80.956 140.564 120.299 210.765 90.900 50.716 250.812 80.631 280.939 90.858 160.709 18
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)
EQ-Net0.743 140.620 800.799 190.849 50.730 160.822 350.493 300.897 40.664 120.681 40.955 180.562 130.378 10.760 110.903 40.738 160.801 110.673 150.907 260.877 50.745 3
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
Virtual MVFusion0.746 110.771 400.819 80.848 60.702 250.865 50.397 700.899 30.699 30.664 70.948 410.588 60.330 90.746 170.851 240.764 90.796 150.704 60.935 110.866 110.728 8
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
SparseConvNet0.725 210.647 760.821 60.846 70.721 190.869 30.533 120.754 430.603 360.614 230.955 180.572 100.325 110.710 210.870 130.724 200.823 20.628 290.934 120.865 120.683 27
StratifiedFormerpermissive0.747 100.901 70.803 170.845 80.757 60.846 140.512 190.825 220.696 50.645 100.956 140.576 90.262 440.744 180.861 160.742 150.770 300.705 50.899 330.860 150.734 6
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
O-CNNpermissive0.762 40.924 20.823 50.844 90.770 20.852 100.577 20.847 150.711 10.640 160.958 100.592 40.217 580.762 100.888 90.758 110.813 70.726 10.932 150.868 90.744 4
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
Mix3Dpermissive0.781 10.964 10.855 10.843 100.781 10.858 70.575 30.831 190.685 70.714 10.979 10.594 30.310 170.801 10.892 80.841 20.819 30.723 30.940 80.887 10.725 12
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
CU-Hybrid Net0.764 20.924 20.819 80.840 110.757 60.853 90.580 10.848 140.709 20.643 120.958 100.587 70.295 230.753 140.884 120.758 110.815 60.725 20.927 170.867 100.743 5
OccuSeg+Semantic0.764 20.758 460.796 200.839 120.746 110.907 10.562 50.850 130.680 90.672 50.978 20.610 10.335 80.777 40.819 320.847 10.830 10.691 80.972 10.885 20.727 10
MinkowskiNetpermissive0.736 190.859 150.818 100.832 130.709 220.840 180.521 180.853 120.660 150.643 120.951 310.544 190.286 280.731 190.893 70.675 390.772 270.683 90.874 510.852 220.727 10
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
JSENetpermissive0.699 290.881 110.762 360.821 140.667 320.800 550.522 170.792 350.613 280.607 270.935 690.492 340.205 630.576 470.853 220.691 340.758 390.652 190.872 540.828 330.649 36
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
PointMetaBase0.714 250.835 210.785 260.821 140.684 290.846 140.531 140.865 100.614 270.596 340.953 260.500 320.246 500.674 230.888 90.692 330.764 330.624 300.849 660.844 300.675 29
Feature-Geometry Netpermissive0.685 340.866 130.748 440.819 160.645 400.794 580.450 470.802 330.587 420.604 280.945 490.464 440.201 660.554 560.840 270.723 210.732 490.602 410.907 260.822 380.603 53
SAT0.742 150.860 140.765 350.819 160.769 30.848 130.533 120.829 200.663 130.631 180.955 180.586 80.274 340.753 140.896 60.729 170.760 370.666 170.921 180.855 200.733 7
PointTransformer++0.725 210.727 600.811 130.819 160.765 40.841 170.502 220.814 290.621 250.623 200.955 180.556 150.284 290.620 380.866 140.781 50.757 400.648 200.932 150.862 130.709 18
RFCR0.702 270.889 90.745 470.813 190.672 310.818 420.493 300.815 270.623 230.610 240.947 430.470 420.249 490.594 420.848 250.705 290.779 240.646 210.892 380.823 360.611 46
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
INS-Conv-semantic0.717 240.751 490.759 380.812 200.704 240.868 40.537 110.842 160.609 320.608 260.953 260.534 210.293 240.616 390.864 150.719 240.793 190.640 240.933 130.845 290.663 32
BPNetcopyleft0.749 70.909 40.818 100.811 210.752 80.839 190.485 320.842 160.673 100.644 110.957 130.528 250.305 190.773 60.859 170.788 40.818 50.693 70.916 210.856 180.723 13
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
MatchingNet0.724 230.812 290.812 120.810 220.735 150.834 240.495 290.860 110.572 480.602 300.954 230.512 290.280 310.757 120.845 260.725 190.780 230.606 390.937 100.851 230.700 21
contrastBoundarypermissive0.705 260.769 430.775 310.809 230.687 280.820 380.439 580.812 300.661 140.591 360.945 490.515 280.171 760.633 340.856 200.720 220.796 150.668 160.889 400.847 260.689 25
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
PointConvFormer0.749 70.793 320.790 240.807 240.750 100.856 80.524 160.881 70.588 410.642 150.977 40.591 50.274 340.781 20.929 10.804 30.796 150.642 230.947 30.885 20.715 17
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
LRPNet0.742 150.816 270.806 150.807 240.752 80.828 290.575 30.839 180.699 30.637 170.954 230.520 270.320 130.755 130.834 280.760 100.772 270.676 120.915 220.862 130.717 15
LargeKernel3D0.739 180.909 40.820 70.806 260.740 130.852 100.545 90.826 210.594 400.643 120.955 180.541 200.263 430.723 200.858 190.775 70.767 310.678 110.933 130.848 240.694 24
DCM-Net0.658 460.778 360.702 620.806 260.619 470.813 480.468 370.693 610.494 680.524 530.941 600.449 530.298 220.510 670.821 310.675 390.727 510.568 550.826 710.803 470.637 40
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
TXC0.740 170.842 190.832 40.805 280.715 210.846 140.473 340.885 60.615 260.671 60.971 60.547 180.320 130.697 220.799 370.777 60.819 30.682 100.946 40.871 80.696 23
DMF-Net0.752 50.906 60.793 230.802 290.689 270.825 310.556 60.867 90.681 80.602 300.960 80.555 160.365 30.779 30.859 170.747 140.795 180.717 40.917 200.856 180.764 2
3DSM_DMMF0.631 600.626 780.745 470.801 300.607 490.751 760.506 200.729 520.565 520.491 640.866 930.434 560.197 690.595 410.630 630.709 280.705 590.560 570.875 490.740 780.491 82
Superpoint Network0.683 380.851 170.728 550.800 310.653 360.806 500.468 370.804 310.572 480.602 300.946 460.453 510.239 530.519 650.822 300.689 370.762 360.595 450.895 360.827 340.630 43
Feature_GeometricNetpermissive0.690 320.884 100.754 420.795 320.647 380.818 420.422 620.802 330.612 290.604 280.945 490.462 450.189 710.563 530.853 220.726 180.765 320.632 270.904 280.821 390.606 50
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
PicassoNet-IIpermissive0.696 300.704 650.790 240.787 330.709 220.837 200.459 420.815 270.543 570.615 220.956 140.529 230.250 470.551 590.790 380.703 300.799 130.619 340.908 250.848 240.700 21
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
PointContrast_LA_SEM0.683 380.757 470.784 270.786 340.639 420.824 330.408 650.775 370.604 350.541 450.934 730.532 220.269 390.552 570.777 390.645 550.793 190.640 240.913 230.824 350.671 30
KP-FCNN0.684 350.847 180.758 400.784 350.647 380.814 450.473 340.772 380.605 340.594 350.935 690.450 520.181 740.587 430.805 350.690 350.785 220.614 350.882 440.819 400.632 42
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
VI-PointConv0.676 400.770 420.754 420.783 360.621 460.814 450.552 70.758 410.571 500.557 410.954 230.529 230.268 410.530 630.682 560.675 390.719 520.603 400.888 410.833 310.665 31
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
DGNet0.684 350.712 640.784 270.782 370.658 330.835 230.499 270.823 240.641 180.597 330.950 350.487 350.281 300.575 480.619 640.647 520.764 330.620 330.871 570.846 280.688 26
VACNN++0.684 350.728 590.757 410.776 380.690 260.804 520.464 400.816 250.577 470.587 370.945 490.508 310.276 330.671 240.710 510.663 440.750 430.589 480.881 450.832 320.653 35
DVVNet0.562 790.648 750.700 640.770 390.586 580.687 840.333 830.650 670.514 650.475 680.906 870.359 790.223 560.340 860.442 810.422 900.668 730.501 770.708 850.779 650.534 75
SALANet0.670 420.816 270.770 330.768 400.652 370.807 490.451 440.747 450.659 160.545 440.924 790.473 410.149 860.571 500.811 340.635 580.746 440.623 310.892 380.794 530.570 63
Retro-FPN0.744 130.842 190.800 180.767 410.740 130.836 220.541 100.914 10.672 110.626 190.958 100.552 170.272 360.777 40.886 110.696 320.801 110.674 140.941 70.858 160.717 15
FusionAwareConv0.630 630.604 830.741 510.766 420.590 550.747 770.501 230.734 500.503 670.527 510.919 830.454 490.323 120.550 600.420 820.678 380.688 660.544 650.896 350.795 520.627 44
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
ROSMRF3D0.673 410.789 330.748 440.763 430.635 440.814 450.407 670.747 450.581 460.573 380.950 350.484 360.271 380.607 400.754 420.649 490.774 260.596 430.883 430.823 360.606 50
SIConv0.625 660.830 230.694 680.757 440.563 650.772 700.448 480.647 690.520 620.509 570.949 390.431 590.191 700.496 710.614 650.647 520.672 720.535 710.876 480.783 640.571 62
FusionNet0.688 330.704 650.741 510.754 450.656 340.829 270.501 230.741 480.609 320.548 430.950 350.522 260.371 20.633 340.756 410.715 260.771 290.623 310.861 620.814 410.658 33
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
SConv0.636 560.830 230.697 660.752 460.572 630.780 660.445 510.716 540.529 600.530 500.951 310.446 550.170 770.507 690.666 600.636 570.682 680.541 680.886 420.799 480.594 57
PointASNLpermissive0.666 430.703 670.781 290.751 470.655 350.830 260.471 360.769 390.474 760.537 470.951 310.475 400.279 320.635 320.698 550.675 390.751 420.553 620.816 730.806 450.703 20
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
One Thing One Click0.701 280.825 250.796 200.723 480.716 200.832 250.433 600.816 250.634 210.609 250.969 70.418 670.344 50.559 540.833 290.715 260.808 90.560 570.902 300.847 260.680 28
PPCNN++permissive0.663 450.746 500.708 590.722 490.638 430.820 380.451 440.566 800.599 380.541 450.950 350.510 300.313 160.648 290.819 320.616 630.682 680.590 470.869 580.810 440.656 34
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
PointConv-SFPN0.641 500.776 380.703 610.721 500.557 670.826 300.451 440.672 650.563 540.483 650.943 570.425 640.162 810.644 300.726 470.659 460.709 560.572 520.875 490.786 630.559 68
DenSeR0.628 640.800 300.625 850.719 510.545 700.806 500.445 510.597 750.448 820.519 560.938 650.481 370.328 100.489 730.499 770.657 470.759 380.592 460.881 450.797 510.634 41
Supervoxel-CNN0.635 570.656 740.711 580.719 510.613 480.757 750.444 540.765 400.534 590.566 390.928 770.478 390.272 360.636 310.531 720.664 430.645 780.508 760.864 610.792 580.611 46
dtc_net0.596 710.683 690.725 560.715 530.549 690.803 530.444 540.647 690.493 690.495 620.941 600.409 690.000 990.424 810.544 690.598 680.703 610.522 730.912 240.792 580.520 78
PointSPNet0.637 550.734 560.692 700.714 540.576 610.797 570.446 490.743 470.598 390.437 760.942 580.403 710.150 850.626 360.800 360.649 490.697 620.557 600.846 670.777 670.563 66
FPConvpermissive0.639 530.785 340.760 370.713 550.603 500.798 560.392 720.534 850.603 360.524 530.948 410.457 470.250 470.538 610.723 490.598 680.696 630.614 350.872 540.799 480.567 65
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
SegGroup_sempermissive0.627 650.818 260.747 460.701 560.602 510.764 720.385 760.629 720.490 710.508 580.931 760.409 690.201 660.564 520.725 480.618 610.692 640.539 690.873 520.794 530.548 72
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
PointConvpermissive0.666 430.781 350.759 380.699 570.644 410.822 350.475 330.779 360.564 530.504 610.953 260.428 610.203 650.586 450.754 420.661 450.753 410.588 490.902 300.813 430.642 38
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
RandLA-Netpermissive0.645 490.778 360.731 540.699 570.577 600.829 270.446 490.736 490.477 750.523 550.945 490.454 490.269 390.484 740.749 450.618 610.738 450.599 420.827 700.792 580.621 45
wsss-transformer0.600 700.634 770.743 490.697 590.601 520.781 640.437 590.585 780.493 690.446 730.933 740.394 730.011 970.654 270.661 620.603 650.733 480.526 720.832 690.761 730.480 84
PointMRNet0.640 520.717 630.701 630.692 600.576 610.801 540.467 390.716 540.563 540.459 710.953 260.429 600.169 780.581 460.854 210.605 640.710 540.550 630.894 370.793 550.575 61
Pointnet++ & Featurepermissive0.557 800.735 550.661 790.686 610.491 770.744 780.392 720.539 830.451 810.375 840.946 460.376 770.205 630.403 830.356 860.553 780.643 790.497 780.824 720.756 740.515 79
CCRFNet0.589 740.766 440.659 800.683 620.470 810.740 790.387 750.620 740.490 710.476 670.922 810.355 810.245 510.511 660.511 750.571 760.643 790.493 800.872 540.762 720.600 54
ROSMRF0.580 750.772 390.707 600.681 630.563 650.764 720.362 790.515 860.465 790.465 700.936 680.427 630.207 610.438 770.577 670.536 790.675 710.486 810.723 840.779 650.524 77
PointMTL0.632 590.731 570.688 730.675 640.591 540.784 630.444 540.565 810.610 300.492 630.949 390.456 480.254 460.587 430.706 520.599 670.665 740.612 380.868 590.791 620.579 60
APCF-Net0.631 600.742 530.687 750.672 650.557 670.792 610.408 650.665 660.545 560.508 580.952 300.428 610.186 720.634 330.702 530.620 600.706 580.555 610.873 520.798 500.581 59
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
PointNet2-SFPN0.631 600.771 400.692 700.672 650.524 720.837 200.440 570.706 590.538 580.446 730.944 550.421 660.219 570.552 570.751 440.591 710.737 460.543 670.901 320.768 700.557 69
MVPNetpermissive0.641 500.831 220.715 570.671 670.590 550.781 640.394 710.679 630.642 170.553 420.937 660.462 450.256 450.649 280.406 830.626 590.691 650.666 170.877 470.792 580.608 49
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
TextureNetpermissive0.566 780.672 730.664 780.671 670.494 760.719 800.445 510.678 640.411 880.396 810.935 690.356 800.225 550.412 820.535 710.565 770.636 820.464 840.794 760.680 880.568 64
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
joint point-basedpermissive0.634 580.614 810.778 300.667 690.633 450.825 310.420 630.804 310.467 780.561 400.951 310.494 330.291 250.566 510.458 780.579 750.764 330.559 590.838 680.814 410.598 55
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
SAFNet-segpermissive0.654 480.752 480.734 530.664 700.583 590.815 440.399 690.754 430.639 190.535 490.942 580.470 420.309 180.665 250.539 700.650 480.708 570.635 260.857 640.793 550.642 38
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
SQN_0.1%0.569 770.676 710.696 670.657 710.497 750.779 670.424 610.548 820.515 640.376 830.902 900.422 650.357 40.379 840.456 790.596 700.659 750.544 650.685 870.665 910.556 70
One-Thing-One-Click0.693 310.743 520.794 220.655 720.684 290.822 350.497 280.719 530.622 240.617 210.977 40.447 540.339 60.750 160.664 610.703 300.790 210.596 430.946 40.855 200.647 37
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
HPGCNN0.656 470.698 680.743 490.650 730.564 640.820 380.505 210.758 410.631 220.479 660.945 490.480 380.226 540.572 490.774 400.690 350.735 470.614 350.853 650.776 680.597 56
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
AttAN0.609 690.760 450.667 770.649 740.521 730.793 590.457 430.648 680.528 610.434 780.947 430.401 720.153 840.454 760.721 500.648 510.717 530.536 700.904 280.765 710.485 83
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
GMLPs0.538 810.495 910.693 690.647 750.471 800.793 590.300 860.477 870.505 660.358 850.903 890.327 840.081 920.472 750.529 730.448 880.710 540.509 740.746 800.737 790.554 71
HPEIN0.618 670.729 580.668 760.647 750.597 530.766 710.414 640.680 620.520 620.525 520.946 460.432 570.215 590.493 720.599 660.638 560.617 830.570 530.897 340.806 450.605 52
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
3DMV0.484 870.484 930.538 940.643 770.424 850.606 950.310 840.574 790.433 860.378 820.796 950.301 870.214 600.537 620.208 940.472 870.507 950.413 930.693 860.602 940.539 73
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PD-Net0.638 540.797 310.769 340.641 780.590 550.820 380.461 410.537 840.637 200.536 480.947 430.388 750.206 620.656 260.668 590.647 520.732 490.585 500.868 590.793 550.473 87
LAP-D0.594 720.720 610.692 700.637 790.456 820.773 690.391 740.730 510.587 420.445 750.940 630.381 760.288 260.434 790.453 800.591 710.649 760.581 510.777 770.749 770.610 48
subcloud_weak0.516 830.676 710.591 920.609 800.442 830.774 680.335 820.597 750.422 870.357 860.932 750.341 830.094 910.298 880.528 740.473 860.676 700.495 790.602 930.721 830.349 94
SPLAT Netcopyleft0.393 950.472 950.511 950.606 810.311 960.656 860.245 930.405 890.328 950.197 970.927 780.227 960.000 990.001 1000.249 900.271 980.510 930.383 960.593 940.699 860.267 96
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
PanopticFusion-label0.529 820.491 920.688 730.604 820.386 870.632 910.225 960.705 600.434 850.293 910.815 940.348 820.241 520.499 700.669 580.507 810.649 760.442 900.796 750.602 940.561 67
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
DPC0.592 730.720 610.700 640.602 830.480 780.762 740.380 770.713 570.585 450.437 760.940 630.369 780.288 260.434 790.509 760.590 730.639 810.567 560.772 780.755 750.592 58
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
PNET20.442 910.548 880.548 930.597 840.363 910.628 930.300 860.292 930.374 900.307 900.881 910.268 920.186 720.238 930.204 950.407 910.506 960.449 870.667 890.620 930.462 88
Online SegFusion0.515 840.607 820.644 830.579 850.434 840.630 920.353 800.628 730.440 830.410 790.762 970.307 860.167 790.520 640.403 840.516 800.565 860.447 880.678 880.701 850.514 80
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
FCPNpermissive0.447 890.679 700.604 910.578 860.380 880.682 850.291 890.106 980.483 740.258 960.920 820.258 930.025 960.231 950.325 870.480 850.560 880.463 850.725 830.666 900.231 98
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
PCNN0.498 860.559 860.644 830.560 870.420 860.711 820.229 940.414 880.436 840.352 870.941 600.324 850.155 830.238 930.387 850.493 820.529 920.509 740.813 740.751 760.504 81
3DWSSS0.425 940.525 890.647 810.522 880.324 940.488 980.077 990.712 580.353 920.401 800.636 990.281 900.176 750.340 860.565 680.175 990.551 890.398 940.370 990.602 940.361 92
ScanNetpermissive0.306 990.203 990.366 980.501 890.311 960.524 970.211 970.002 1000.342 940.189 980.786 960.145 990.102 900.245 920.152 970.318 970.348 980.300 980.460 980.437 990.182 99
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
SPH3D-GCNpermissive0.610 680.858 160.772 320.489 900.532 710.792 610.404 680.643 710.570 510.507 600.935 690.414 680.046 950.510 670.702 530.602 660.705 590.549 640.859 630.773 690.534 75
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
Tangent Convolutionspermissive0.438 930.437 960.646 820.474 910.369 890.645 890.353 800.258 950.282 970.279 920.918 840.298 880.147 870.283 900.294 880.487 830.562 870.427 920.619 920.633 920.352 93
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
DGCNN_reproducecopyleft0.446 900.474 940.623 860.463 920.366 900.651 880.310 840.389 910.349 930.330 880.937 660.271 910.126 880.285 890.224 920.350 950.577 850.445 890.625 910.723 820.394 90
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
PointNet++permissive0.339 970.584 840.478 970.458 930.256 980.360 990.250 910.247 960.278 980.261 950.677 980.183 970.117 890.212 970.145 980.364 930.346 990.232 990.548 950.523 980.252 97
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SD-DETR0.576 760.746 500.609 890.445 940.517 740.643 900.366 780.714 560.456 800.468 690.870 920.432 570.264 420.558 550.674 570.586 740.688 660.482 820.739 820.733 800.537 74
ScanNet+FTSDF0.383 960.297 980.491 960.432 950.358 920.612 940.274 900.116 970.411 880.265 940.904 880.229 950.079 930.250 910.185 960.320 960.510 930.385 950.548 950.597 970.394 90
3DMV, FTSDF0.501 850.558 870.608 900.424 960.478 790.690 830.246 920.586 770.468 770.450 720.911 850.394 730.160 820.438 770.212 930.432 890.541 910.475 830.742 810.727 810.477 85
SurfaceConvPF0.442 910.505 900.622 870.380 970.342 930.654 870.227 950.397 900.367 910.276 930.924 790.240 940.198 680.359 850.262 890.366 920.581 840.435 910.640 900.668 890.398 89
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
PointCNN with RGBpermissive0.458 880.577 850.611 880.356 980.321 950.715 810.299 880.376 920.328 950.319 890.944 550.285 890.164 800.216 960.229 910.484 840.545 900.456 860.755 790.709 840.475 86
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
SSC-UNetpermissive0.308 980.353 970.290 990.278 990.166 990.553 960.169 980.286 940.147 990.148 990.908 860.182 980.064 940.023 990.018 1000.354 940.363 970.345 970.546 970.685 870.278 95
ERROR0.054 1000.000 1000.041 1000.172 1000.030 1000.062 1000.001 1000.035 990.004 1000.051 1000.143 1000.019 1000.003 980.041 980.050 990.003 1000.054 1000.018 1000.005 1000.264 1000.082 100


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




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
PBNetpermissive0.573 20.926 20.575 100.619 10.472 20.736 40.239 50.487 250.383 20.459 30.506 60.533 70.585 60.767 70.404 80.717 30.559 50.969 20.381 4
IPCA-Inst0.520 80.889 60.551 140.548 20.418 50.665 180.064 190.585 100.260 170.277 190.471 110.500 90.644 40.785 40.369 90.591 180.511 70.878 180.362 8
DKNet0.532 70.815 90.624 30.517 30.377 90.749 10.107 100.509 220.304 90.437 60.475 90.581 30.539 90.775 60.339 130.640 110.506 90.901 100.385 3
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
SSEC0.465 160.667 270.578 80.502 40.362 150.641 250.035 280.605 60.291 150.323 150.451 140.296 170.417 250.677 200.245 260.501 340.506 100.900 110.366 5
SSTNetpermissive0.506 100.738 170.549 150.497 50.316 190.693 120.178 70.377 330.198 220.330 120.463 130.576 40.515 110.857 20.494 10.637 120.457 150.943 40.290 23
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
SPFormerpermissive0.549 60.745 140.640 20.484 60.395 60.739 20.311 10.566 130.335 60.468 20.492 80.555 60.478 140.747 120.436 60.712 40.540 60.893 130.343 12
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
DD-UNet+Group0.436 180.630 340.508 260.480 70.310 200.624 290.065 180.638 50.174 240.256 230.384 210.194 310.428 210.759 80.289 180.574 220.400 230.849 270.291 22
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
Mask-Group0.434 200.778 120.516 210.471 80.330 170.658 190.029 300.526 200.249 180.256 220.400 190.309 160.384 290.296 460.368 100.575 210.425 200.877 190.362 9
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
TopoSeg0.479 140.704 200.564 110.467 90.366 120.633 260.068 170.554 160.262 160.328 130.447 150.323 130.534 100.722 140.288 190.614 130.482 130.912 60.358 10
HAISpermissive0.457 170.704 200.561 120.457 100.364 130.673 150.046 270.547 170.194 230.308 160.426 160.288 180.454 170.711 160.262 230.563 250.434 190.889 150.344 11
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
Dyco3Dcopyleft0.395 270.642 320.518 200.447 110.259 280.666 170.050 250.251 460.166 250.231 270.362 230.232 260.331 320.535 290.229 270.587 190.438 180.850 250.317 16
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
PE0.396 260.667 270.467 310.446 120.243 300.624 300.022 340.577 110.106 340.219 290.340 250.239 240.487 130.475 370.225 280.541 300.350 300.818 330.273 26
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
GraphCut0.552 51.000 10.611 40.438 130.392 70.714 80.139 80.598 90.327 70.389 70.510 50.598 10.427 230.754 100.463 40.761 10.588 20.903 90.329 14
TD3D0.489 120.852 70.511 230.434 140.322 180.735 50.101 130.512 210.355 50.349 110.468 120.283 190.514 120.676 210.268 220.671 70.510 80.908 70.329 15
Box2Mask0.433 210.741 150.463 330.433 150.283 220.625 280.103 110.298 420.125 320.260 210.424 170.322 140.472 150.701 180.363 110.711 50.309 400.882 160.272 27
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
OccuSeg+instance0.486 130.802 110.536 170.428 160.369 110.702 100.205 60.331 380.301 110.379 90.474 100.327 120.437 190.862 10.485 20.601 160.394 250.846 290.273 25
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
PointGroup0.407 240.639 330.496 270.415 170.243 310.645 240.021 350.570 120.114 330.211 340.359 240.217 290.428 220.660 220.256 240.562 260.341 320.860 230.291 21
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]
Mask3D0.566 30.926 20.597 50.408 180.420 40.737 30.239 40.598 80.386 10.458 40.549 10.568 50.716 20.601 250.480 30.646 90.575 30.922 50.364 6
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
DualGroup0.469 150.815 90.552 130.398 190.374 100.683 140.130 90.539 180.310 80.327 140.407 180.276 200.447 180.535 300.342 120.659 80.455 160.900 120.301 19
SoftGroup++0.513 90.704 200.578 90.398 200.363 140.704 90.061 200.647 40.297 140.378 100.537 20.343 110.614 50.828 30.295 170.710 60.505 110.875 200.394 1
Queryformer0.583 10.926 20.702 10.393 210.504 10.733 60.276 20.527 190.373 40.479 10.534 30.533 80.697 30.720 150.436 70.745 20.592 10.958 30.363 7
ISBNetpermissive0.559 40.926 20.597 60.390 220.436 30.722 70.276 30.556 150.380 30.450 50.505 70.583 20.730 10.575 260.455 50.603 150.573 40.979 10.332 13
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
AOIA0.387 290.704 200.515 220.385 230.225 340.669 160.005 430.482 260.126 310.181 390.269 350.221 280.426 240.478 360.218 290.592 170.371 270.851 240.242 31
NeuralBF0.353 340.593 360.511 240.375 240.264 260.597 310.008 390.332 370.160 260.229 280.274 340.000 560.206 400.678 190.155 400.485 360.422 210.816 340.254 30
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
SoftGrouppermissive0.504 110.667 270.579 70.372 250.381 80.694 110.072 160.677 20.303 100.387 80.531 40.319 150.582 70.754 90.318 140.643 100.492 120.907 80.388 2
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
RPGN0.428 220.630 340.508 250.367 260.249 290.658 200.016 370.673 30.131 300.234 260.383 220.270 210.434 200.748 110.274 210.609 140.406 220.842 310.267 28
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
INS-Conv-instance0.435 190.716 190.495 280.355 270.331 160.689 130.102 120.394 320.208 210.280 170.395 200.250 230.544 80.741 130.309 160.536 310.391 260.842 320.258 29
SphereSeg0.357 320.651 310.411 350.345 280.264 270.630 270.059 210.289 440.212 190.240 250.336 260.158 350.305 330.557 270.159 390.455 400.341 330.726 410.294 20
GICN0.341 360.580 370.371 370.344 290.198 360.469 390.052 230.564 140.093 350.212 320.212 400.127 380.347 310.537 280.206 300.525 320.329 350.729 400.241 32
Occipital-SCS0.320 380.679 260.352 390.334 300.229 320.436 400.025 310.412 300.058 410.161 440.240 370.085 400.262 350.496 350.187 340.467 380.328 360.775 350.231 33
CSC-Pretrained0.405 250.738 170.465 320.331 310.205 350.655 210.051 240.601 70.092 360.211 350.329 270.198 300.459 160.775 50.195 330.524 330.400 240.878 170.184 37
RWSeg0.348 350.475 420.456 340.320 320.275 250.476 380.020 360.491 240.056 430.212 330.320 280.261 220.302 340.520 310.182 350.557 270.285 420.867 210.197 35
3D-MPA0.355 330.457 450.484 300.299 330.277 240.591 320.047 260.332 360.212 200.217 300.278 320.193 320.413 260.410 400.195 320.574 230.352 290.849 260.213 34
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
SPG_WSIS0.251 440.380 470.274 460.289 340.144 410.413 430.000 490.311 400.065 390.113 460.130 460.029 480.204 410.388 410.108 460.459 390.311 380.769 360.127 44
PCJC0.375 310.704 200.542 160.284 350.197 370.649 220.006 410.426 270.138 280.242 240.304 310.183 340.388 280.629 230.141 430.546 290.344 310.738 390.283 24
PanopticFusion-inst0.214 460.250 510.330 420.275 360.103 470.228 550.000 490.345 350.024 470.088 480.203 420.186 330.167 450.367 430.125 440.221 530.112 550.666 440.162 40
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-BoNet0.253 430.519 400.324 430.251 370.137 430.345 490.031 290.419 290.069 380.162 430.131 450.052 440.202 420.338 440.147 420.301 500.303 410.651 450.178 38
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
DENet0.413 230.741 150.520 190.237 380.284 210.523 360.097 140.691 10.138 270.209 360.229 380.238 250.390 270.707 170.310 150.448 410.470 140.892 140.310 17
One_Thing_One_Clickpermissive0.326 370.472 430.361 380.232 390.183 380.555 340.000 490.498 230.038 450.195 370.226 390.362 100.168 440.469 380.251 250.553 280.335 340.846 280.117 45
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
OSIS0.392 280.778 120.530 180.220 400.278 230.567 330.083 150.330 390.299 120.270 200.310 300.143 360.260 360.624 240.277 200.568 240.361 280.865 220.301 18
SSEN0.384 300.852 70.494 290.192 410.226 330.648 230.022 330.398 310.299 130.277 180.317 290.231 270.194 430.514 330.196 310.586 200.444 170.843 300.184 36
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
MTML0.282 400.577 380.380 360.182 420.107 460.430 410.001 460.422 280.057 420.179 400.162 430.070 430.229 380.511 340.161 370.491 350.313 370.650 460.162 39
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
Sparse R-CNN0.292 390.704 200.213 490.153 430.154 400.551 350.053 220.212 470.132 290.174 410.274 330.070 420.363 300.441 390.176 360.424 430.234 440.758 370.161 41
MASCpermissive0.254 420.463 440.249 480.113 440.167 390.412 440.000 480.374 340.073 370.173 420.243 360.130 370.228 390.368 420.160 380.356 450.208 450.711 430.136 43
Chen Liu, Yasutaka Furukawa: MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation.
R-PointNet0.158 490.356 480.173 510.113 450.140 420.359 450.012 380.023 540.039 440.134 450.123 470.008 520.089 490.149 510.117 450.221 520.128 520.563 480.094 46
UNet-backbone0.161 470.519 400.259 470.084 460.059 490.325 510.002 440.093 520.009 510.077 500.064 490.045 450.044 530.161 500.045 490.331 480.180 470.566 470.033 56
Region-18class0.146 500.175 550.321 440.080 470.062 480.357 460.000 490.307 410.002 530.066 510.044 510.000 560.018 550.036 550.054 480.447 420.133 500.472 490.060 51
SemRegionNet-20cls0.121 510.296 500.203 500.071 480.058 500.349 470.000 490.150 480.019 480.054 520.034 530.017 510.052 510.042 540.013 550.209 540.183 460.371 510.057 52
3D-SISpermissive0.161 470.407 460.155 530.068 490.043 530.346 480.001 450.134 490.005 520.088 470.106 480.037 470.135 480.321 450.028 520.339 460.116 540.466 500.093 47
Ji Hou, Angela Dai, Matthias Niessner: 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans. CVPR 2019
SALoss-ResNet0.262 410.667 270.335 400.067 500.123 440.427 420.022 320.280 450.058 400.216 310.211 410.039 460.142 460.519 320.106 470.338 470.310 390.721 420.138 42
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)
ASIS0.085 550.037 560.080 560.066 510.047 520.282 530.000 490.052 530.002 540.047 540.026 540.001 550.046 520.194 490.031 510.264 510.140 490.167 550.047 55
SegGroup_inspermissive0.246 450.556 390.335 410.062 520.115 450.490 370.000 490.297 430.018 490.186 380.142 440.083 410.233 370.216 480.153 410.469 370.251 430.744 380.083 48
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
tmp0.113 540.333 490.151 540.056 530.053 510.344 500.000 490.105 510.016 500.049 530.035 520.020 500.053 500.048 530.013 540.183 550.173 480.344 530.054 53
Hier3Dcopyleft0.117 520.222 530.161 520.054 540.027 540.289 520.000 490.124 500.001 550.079 490.061 500.027 490.141 470.240 470.005 560.310 490.129 510.153 560.081 49
Tan: HCFS3D: Hierarchical Coupled Feature Selection Network for 3D Semantic and Instance Segmentation.
Sgpn_scannet0.049 560.023 570.134 550.031 550.013 560.144 560.006 400.008 560.000 560.028 560.017 550.003 530.009 570.000 560.021 530.122 560.095 560.175 540.054 54
3D-BEVIS0.117 520.250 510.308 450.020 560.009 570.269 540.006 420.008 550.029 460.037 550.014 560.003 540.036 540.147 520.042 500.381 440.118 530.362 520.069 50
Cathrin Elich, Francis Engelmann, Jonas Schult, Theodora Kontogianni, Bastian Leibe: 3D-BEVIS: Birds-Eye-View Instance Segmentation.
MaskRCNN 2d->3d Proj0.022 570.185 540.000 570.000 570.015 550.000 570.000 470.006 570.000 560.010 570.006 570.107 390.012 560.000 560.002 570.027 570.004 570.022 570.001 57


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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort 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
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)
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
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
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
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
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
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
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
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
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
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
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.
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
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
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
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
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
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
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
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
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
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
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 bysorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort 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