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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
3DMV0.443 10.328 20.510 20.418 40.353 30.595 30.252 30.681 10.398 20.365 10.790 30.286 20.215 10.481 10.195 20.371 20.491 30.321 30.669 10.610 20.523 1
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
Tangent Convolutionspermissive0.438 20.437 10.646 10.474 20.369 10.645 10.353 10.258 20.282 40.279 20.918 10.298 10.147 20.283 20.294 10.487 10.562 10.427 10.619 20.633 10.352 3
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
ScanNet+FTSDF0.383 30.297 30.491 30.432 30.358 20.612 20.274 20.116 30.411 10.265 30.904 20.229 30.079 40.250 30.185 30.320 30.510 20.385 20.548 30.597 30.394 2
ScanNetpermissive0.306 40.203 40.366 40.501 10.311 40.524 40.211 40.002 40.342 30.189 40.786 40.145 40.102 30.245 40.152 40.318 40.348 40.300 40.460 40.437 40.182 4
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17

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




Method Infoavg ap 25%bathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
MaskRCNN 2d->3d Proj0.227 10.850 10.074 10.002 10.191 10.150 10.221 10.103 10.073 10.131 10.147 10.387 10.197 10.143 10.532 10.356 10.117 10.380 10.030 1

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


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
ILC-PSPNet0.475 10.490 10.581 10.289 40.507 10.067 40.379 10.610 20.417 30.435 10.822 30.278 10.267 20.503 20.228 10.616 10.533 20.375 10.820 10.729 10.560 1
3DMV (2d proj)0.453 20.379 20.578 20.290 30.376 20.335 10.296 20.676 10.523 10.395 20.839 10.220 20.281 10.505 10.182 20.452 20.593 10.366 20.689 20.668 30.408 3
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
Enet (reimpl)0.376 30.264 40.452 40.452 20.365 30.181 20.143 40.456 30.409 40.346 30.769 40.164 30.218 30.359 30.123 40.403 40.381 40.313 40.571 30.685 20.472 2
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 40.293 30.521 30.657 10.361 40.161 30.250 30.004 40.440 20.183 40.836 20.125 40.060 40.319 40.132 30.417 30.412 30.344 30.541 40.427 40.109 4
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17

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




Method Infoavg apbathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
MaskRCNN_ScanNetpermissive0.119 10.129 10.212 10.002 10.112 10.148 10.014 10.205 10.044 10.066 10.078 10.095 10.142 10.030 10.128 10.139 10.080 10.459 10.057 1
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
resnet50_scannet0.353 10.250 10.812 10.529 10.500 10.500 10.000 10.500 10.571 10.000 10.556 10.000 10.375 10.000 1