Benchmark Results - ScanNet Benchmark

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

Method	avg iou	head iou	common iou	tail iou	wall	chair	floor	table	door	couch	cabinet	shelf	desk	office chair	bed	pillow	sink	picture	window	toilet	bookshelf	monitor	curtain	book	coffee table	box	refrigerator	lamp	kitchen cabinet	towel	clothes	tv	nightstand	counter	dresser	stool	cushion	plant	ceiling	bathtub	end table	dining table	keyboard	bag	backpack	toilet paper	printer	tv stand	whiteboard	blanket	shower curtain	trash can	closet	stairs	microwave	stove	shoe	computer tower	bottle	bin	bench	board	washing machine	mirror	copier	basket	sofa chair	file cabinet	fan	laptop	shower	paper	person	paper towel dispenser	oven	blinds	rack	plate	blackboard	piano	suitcase	radiator	recycling bin	wardrobe	soap dispenser	telephone	bucket	clock	stand	light	laundry basket	pipe	clothes dryer	seat	speaker	column	bicycle	ladder	bathroom stall	shower wall	cup	jacket	storage bin	coffee maker	dishwasher	paper towel roll	machine	mat	windowsill	bar	toaster	bulletin board	ironing board	kitchen counter	doorframe	toilet paper dispenser	mini fridge	fire extinguisher	ball	hat	shower curtain rod	water cooler	paper cutter	tray	ledge	mouse	cart	storage container	scale	tissue box	light switch	power outlet	decoration	sign	projector	vacuum cleaner	candle	plunger	stuffed animal	headphones	dish rack	broom	range hood	water bottle	vent	shower floor	water pitcher	mailbox	bowl	paper bag	music stand	projector screen	laundry detergent	object	bathroom vanity	laundry hamper	bathroom stall door	ceiling light	trash bin	dumbbell	stair rail	tube	bathroom cabinet	closet rod	coffee kettle	structure	shower head	keyboard piano	case of water bottles	coat rack	storage organizer	folded chair	fire alarm	power strip	calendar	poster

OctFormer ScanNet200	0.326 8	0.539 7	0.265 7	0.131 7	0.806 6	0.670 8	0.943 7	0.535 7	0.662 2	0.705 11	0.423 6	0.407 3	0.505 8	0.003 8	0.765 8	0.582 5	0.686 10	0.227 11	0.680 5	0.943 6	0.601 1	0.854 8	0.892 3	0.335 3	0.417 12	0.357 7	0.724 7	0.453 6	0.632 4	0.596 2	0.432 3	0.783 6	0.512 11	0.021 10	0.244 10	0.637 1	0.000 1	0.787 7	0.873 7	0.743 10	0.000 12	0.000 7	0.534 6	0.110 1	0.499 4	0.289 5	0.626 4	0.620 9	0.168 12	0.204 1	0.849 5	0.679 5	0.117 3	0.633 7	0.684 2	0.650 5	0.552 3	0.684 7	0.312 2	0.000 3	0.175 7	0.429 7	0.865 3	0.413 2	0.837 7	0.000 3	0.145 6	0.626 6	0.451 5	0.487 8	0.513 1	0.000 1	0.529 5	0.613 8	0.000 4	0.033 3	0.000 1	0.000 3	0.828 2	0.871 2	0.622 6	0.587 6	0.411 5	0.137 9	0.645 9	0.343 7	0.000 3	0.000 5	0.000 1	0.022 8	0.000 3	0.026 12	0.829 8	0.000 1	0.022 5	0.089 4	0.842 1	0.253 11	0.318 8	0.296 2	0.178 6	0.291 3	0.224 1	0.584 2	0.200 9	0.132 6	0.000 3	0.128 6	0.227 11	0.000 1	0.230 8	0.047 8	0.149 5	0.331 7	0.412 7	0.618 5	0.164 5	0.102 6	0.522 1	0.000 1	0.655 3	0.378 8	0.469 10	0.000 1	0.000 7	0.000 7	0.105 6	0.000 5	0.000 7	0.483 3	0.000 7	0.000 4	0.028 5	0.000 1	0.000 1	0.906 1	0.000 1	0.339 10	0.000 1	0.000 8	0.457 7	0.000 1	0.612 5	0.000 1	0.000 1	0.408 2	0.000 10	0.900 6	0.000 5	0.000 6	0.000 1	0.029 4	0.000 1	0.074 12	0.455 10	0.479 4	0.427 4	0.079 8	0.140 8	0.496 6	0.414 9	0.022 2	0.000 1	0.471 9	0.000 2	0.000 2	0.000 7	0.722 4	0.000 2	0.000 1	0.000 1	0.138 9	0.000 4	0.000 2	0.000 6	0.000 1
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
PPT-SpUNet-F.T.	0.332 7	0.556 3	0.270 4	0.123 9	0.816 4	0.682 5	0.946 4	0.549 5	0.657 6	0.756 3	0.459 4	0.376 6	0.550 6	0.001 9	0.807 3	0.616 2	0.727 7	0.267 4	0.691 3	0.942 7	0.530 7	0.872 4	0.874 5	0.330 5	0.542 9	0.374 5	0.792 3	0.400 9	0.673 2	0.572 4	0.433 2	0.793 4	0.623 4	0.008 12	0.351 5	0.594 6	0.000 1	0.783 8	0.876 5	0.833 4	0.213 3	0.000 7	0.537 5	0.091 4	0.519 2	0.304 4	0.620 5	0.942 1	0.264 3	0.124 5	0.855 4	0.695 3	0.086 6	0.646 6	0.506 11	0.658 4	0.535 4	0.715 2	0.314 1	0.000 3	0.241 3	0.608 2	0.897 2	0.359 6	0.858 6	0.000 3	0.076 12	0.611 8	0.392 7	0.509 6	0.378 3	0.000 1	0.579 2	0.565 11	0.000 4	0.000 6	0.000 1	0.000 3	0.755 4	0.806 8	0.661 2	0.572 10	0.350 7	0.181 7	0.660 7	0.300 9	0.000 3	0.000 5	0.000 1	0.023 7	0.000 3	0.042 11	0.930 2	0.000 1	0.000 7	0.077 5	0.584 4	0.392 7	0.339 6	0.185 5	0.171 7	0.308 2	0.006 10	0.563 3	0.256 6	0.150 2	0.000 3	0.002 11	0.345 10	0.000 1	0.045 9	0.197 2	0.063 6	0.323 8	0.453 3	0.600 6	0.163 6	0.037 10	0.349 2	0.000 1	0.672 2	0.679 3	0.753 1	0.000 1	0.000 7	0.000 7	0.117 3	0.000 5	0.000 7	0.291 8	0.000 7	0.000 4	0.039 4	0.000 1	0.000 1	0.899 2	0.000 1	0.374 8	0.000 1	0.000 8	0.545 4	0.000 1	0.634 3	0.000 1	0.000 1	0.074 8	0.223 3	0.914 5	0.000 5	0.021 4	0.000 1	0.000 5	0.000 1	0.112 4	0.498 8	0.649 1	0.383 7	0.095 2	0.135 11	0.449 8	0.432 7	0.008 5	0.000 1	0.518 5	0.000 2	0.000 2	0.000 7	0.796 3	0.000 2	0.000 1	0.000 1	0.138 9	0.000 4	0.000 2	0.000 6	0.000 1
Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao: Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training. CVPR 2024
OA-CNN-L_ScanNet200	0.333 6	0.558 2	0.269 6	0.124 8	0.821 3	0.703 1	0.946 4	0.569 2	0.662 2	0.748 6	0.487 2	0.455 1	0.572 4	0.000 10	0.789 5	0.534 6	0.736 6	0.271 3	0.713 2	0.949 4	0.498 11	0.877 2	0.860 6	0.332 4	0.706 1	0.474 1	0.788 5	0.406 8	0.637 3	0.495 6	0.355 7	0.805 3	0.592 9	0.015 11	0.396 3	0.602 5	0.000 1	0.799 6	0.876 5	0.713 12	0.276 1	0.000 7	0.493 8	0.080 6	0.448 10	0.363 2	0.661 2	0.833 5	0.262 4	0.125 4	0.823 7	0.665 6	0.076 7	0.720 4	0.557 6	0.637 6	0.517 6	0.672 8	0.227 6	0.000 3	0.158 8	0.496 4	0.843 8	0.352 7	0.835 8	0.000 3	0.103 10	0.711 3	0.527 2	0.526 5	0.320 5	0.000 1	0.568 4	0.625 7	0.067 1	0.000 6	0.000 1	0.001 2	0.806 3	0.836 6	0.621 7	0.591 5	0.373 6	0.314 5	0.668 5	0.398 6	0.003 2	0.000 5	0.000 1	0.016 11	0.024 2	0.043 10	0.906 4	0.000 1	0.052 4	0.000 8	0.384 7	0.330 9	0.342 5	0.100 7	0.223 4	0.183 8	0.112 4	0.476 5	0.313 5	0.130 7	0.196 2	0.112 7	0.370 9	0.000 1	0.234 7	0.071 6	0.160 4	0.403 3	0.398 9	0.492 10	0.197 2	0.076 8	0.272 3	0.000 1	0.200 12	0.560 6	0.735 3	0.000 1	0.000 7	0.000 7	0.110 5	0.002 4	0.021 6	0.412 5	0.000 7	0.000 4	0.000 8	0.000 1	0.000 1	0.794 7	0.000 1	0.445 3	0.000 1	0.022 6	0.509 6	0.000 1	0.517 10	0.000 1	0.000 1	0.001 12	0.245 2	0.915 4	0.024 2	0.089 3	0.000 1	0.262 2	0.000 1	0.103 8	0.524 5	0.392 8	0.515 2	0.013 12	0.251 4	0.411 10	0.662 1	0.001 7	0.000 1	0.473 8	0.000 2	0.000 2	0.150 4	0.699 6	0.000 2	0.000 1	0.000 1	0.166 5	0.000 4	0.024 1	0.000 6	0.000 1

BFANet ScanNet200	0.360 2	0.553 4	0.293 2	0.193 2	0.827 2	0.689 2	0.970 2	0.528 8	0.661 4	0.753 4	0.436 5	0.378 5	0.469 10	0.042 5	0.810 2	0.654 1	0.760 2	0.266 5	0.659 7	0.973 2	0.574 2	0.849 9	0.897 2	0.382 1	0.546 8	0.372 6	0.698 9	0.491 5	0.617 5	0.526 5	0.436 1	0.764 9	0.476 12	0.101 5	0.409 2	0.585 7	0.000 1	0.835 1	0.901 2	0.810 5	0.102 9	0.000 7	0.688 1	0.096 3	0.483 6	0.264 7	0.612 6	0.591 11	0.358 1	0.161 3	0.863 3	0.707 2	0.128 2	0.814 1	0.669 3	0.629 7	0.563 2	0.651 10	0.258 3	0.000 3	0.194 6	0.494 5	0.806 9	0.394 4	0.953 2	0.000 3	0.233 1	0.757 2	0.508 4	0.556 3	0.476 2	0.000 1	0.573 3	0.741 3	0.000 4	0.000 6	0.000 1	0.000 3	0.000 12	0.852 4	0.678 1	0.616 3	0.460 3	0.338 3	0.710 2	0.534 2	0.000 3	0.025 2	0.000 1	0.043 2	0.000 3	0.056 9	0.493 12	0.000 1	0.000 7	0.109 3	0.785 2	0.590 3	0.298 10	0.282 3	0.143 8	0.262 4	0.053 8	0.526 4	0.337 3	0.215 1	0.000 3	0.135 5	0.510 3	0.000 1	0.596 1	0.043 9	0.511 2	0.321 9	0.459 2	0.772 1	0.124 8	0.060 9	0.266 4	0.000 1	0.574 6	0.568 5	0.653 6	0.000 1	0.093 1	0.298 2	0.239 1	0.000 5	0.516 2	0.129 9	0.284 1	0.000 4	0.431 1	0.000 1	0.000 1	0.848 6	0.000 1	0.492 1	0.000 1	0.376 2	0.522 5	0.000 1	0.469 12	0.000 1	0.000 1	0.330 4	0.151 5	0.875 10	0.000 5	0.254 1	0.000 1	0.000 5	0.000 1	0.088 10	0.661 1	0.481 3	0.255 9	0.105 1	0.139 9	0.666 3	0.641 2	0.000 8	0.000 1	0.614 1	0.000 2	0.000 2	0.000 7	0.921 1	0.000 2	0.000 1	0.000 1	0.497 1	0.000 4	0.000 2	0.000 6	0.000 1

PonderV2 ScanNet200	0.346 3	0.552 5	0.270 5	0.175 4	0.810 5	0.682 5	0.950 3	0.560 4	0.641 7	0.761 1	0.398 8	0.357 7	0.570 5	0.113 2	0.804 4	0.603 4	0.750 4	0.283 2	0.681 4	0.952 3	0.548 3	0.874 3	0.852 8	0.290 7	0.700 2	0.356 8	0.792 3	0.445 7	0.545 8	0.436 7	0.351 8	0.787 5	0.611 5	0.050 7	0.290 9	0.519 9	0.000 1	0.825 5	0.888 3	0.842 3	0.259 2	0.100 2	0.558 4	0.070 9	0.497 5	0.247 9	0.457 8	0.889 2	0.248 6	0.106 7	0.817 8	0.691 4	0.094 5	0.729 2	0.636 4	0.620 9	0.503 8	0.660 9	0.243 5	0.000 3	0.212 5	0.590 3	0.860 6	0.400 3	0.881 4	0.000 3	0.202 2	0.622 7	0.408 6	0.499 7	0.261 7	0.000 1	0.385 6	0.636 6	0.000 4	0.000 6	0.000 1	0.000 3	0.433 11	0.843 5	0.660 4	0.574 9	0.481 2	0.336 4	0.677 4	0.486 3	0.000 3	0.030 1	0.000 1	0.034 5	0.000 3	0.080 5	0.869 7	0.000 1	0.000 7	0.000 8	0.540 5	0.727 2	0.232 12	0.115 6	0.186 5	0.193 6	0.000 11	0.403 7	0.326 4	0.103 9	0.000 3	0.290 3	0.392 7	0.000 1	0.346 5	0.062 7	0.424 3	0.375 4	0.431 4	0.667 3	0.115 9	0.082 7	0.239 5	0.000 1	0.504 9	0.606 4	0.584 7	0.000 1	0.002 5	0.186 5	0.104 7	0.000 5	0.394 3	0.384 6	0.083 5	0.000 4	0.007 6	0.000 1	0.000 1	0.880 4	0.000 1	0.377 7	0.000 1	0.263 3	0.565 2	0.000 1	0.608 6	0.000 1	0.000 1	0.304 5	0.009 6	0.924 1	0.000 5	0.000 6	0.000 1	0.000 5	0.000 1	0.128 2	0.584 2	0.475 5	0.412 5	0.076 9	0.269 3	0.621 4	0.509 4	0.010 3	0.000 1	0.491 7	0.063 1	0.000 2	0.472 3	0.880 2	0.000 2	0.000 1	0.000 1	0.179 4	0.125 1	0.000 2	0.441 5	0.000 1
Haoyi Zhu, Honghui Yang, Xiaoyang Wu, Di Huang, Sha Zhang, Xianglong He, Tong He, Hengshuang Zhao, Chunhua Shen, Yu Qiao, Wanli Ouyang: PonderV2: Pave the Way for 3D Foundataion Model with A Universal Pre-training Paradigm.
LGround	0.272 10	0.485 10	0.184 10	0.106 10	0.778 10	0.676 7	0.932 10	0.479 12	0.572 10	0.718 9	0.399 7	0.265 10	0.453 11	0.085 3	0.745 10	0.446 10	0.726 8	0.232 10	0.622 10	0.901 10	0.512 9	0.826 10	0.786 11	0.178 11	0.549 7	0.277 10	0.659 10	0.381 10	0.518 9	0.295 12	0.323 9	0.777 7	0.599 7	0.028 8	0.321 6	0.363 11	0.000 1	0.708 10	0.858 10	0.746 9	0.063 10	0.022 5	0.457 10	0.077 7	0.476 7	0.243 10	0.402 9	0.397 12	0.233 7	0.077 12	0.720 12	0.610 11	0.103 4	0.629 8	0.437 12	0.626 8	0.446 9	0.702 4	0.190 8	0.005 1	0.058 11	0.322 10	0.702 11	0.244 10	0.768 9	0.000 3	0.134 8	0.552 10	0.279 11	0.395 10	0.147 11	0.000 1	0.207 10	0.612 9	0.000 4	0.000 6	0.000 1	0.000 3	0.658 6	0.566 10	0.323 10	0.525 12	0.229 9	0.179 8	0.467 12	0.154 11	0.000 3	0.002 3	0.000 1	0.051 1	0.000 3	0.127 2	0.703 9	0.000 1	0.000 7	0.216 1	0.112 11	0.358 8	0.547 1	0.187 4	0.092 11	0.156 12	0.055 7	0.296 10	0.252 7	0.143 3	0.000 3	0.014 9	0.398 6	0.000 1	0.028 11	0.173 4	0.000 10	0.265 11	0.348 10	0.415 11	0.179 3	0.019 11	0.218 6	0.000 1	0.597 5	0.274 12	0.565 8	0.000 1	0.012 4	0.000 7	0.039 11	0.022 2	0.000 7	0.117 10	0.000 7	0.000 4	0.000 8	0.000 1	0.000 1	0.324 11	0.000 1	0.384 6	0.000 1	0.000 8	0.251 12	0.000 1	0.566 8	0.000 1	0.000 1	0.066 9	0.404 1	0.886 9	0.199 1	0.000 6	0.000 1	0.059 3	0.000 1	0.136 1	0.540 4	0.127 12	0.295 8	0.085 6	0.143 6	0.514 5	0.413 10	0.000 8	0.000 1	0.498 6	0.000 2	0.000 2	0.000 7	0.623 8	0.000 2	0.000 1	0.000 1	0.132 11	0.000 4	0.000 2	0.000 6	0.000 1
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild. arXiv
AWCS	0.305 9	0.508 9	0.225 9	0.142 6	0.782 9	0.634 12	0.937 9	0.489 10	0.578 9	0.721 7	0.364 10	0.355 8	0.515 7	0.023 7	0.764 9	0.523 8	0.707 9	0.264 6	0.633 9	0.922 9	0.507 10	0.886 1	0.804 10	0.179 10	0.436 11	0.300 9	0.656 11	0.529 2	0.501 10	0.394 8	0.296 11	0.820 2	0.603 6	0.131 3	0.179 12	0.619 2	0.000 1	0.707 11	0.865 9	0.773 6	0.171 4	0.010 6	0.484 9	0.063 10	0.463 9	0.254 8	0.332 11	0.649 8	0.220 8	0.100 8	0.729 10	0.613 10	0.071 9	0.582 10	0.628 5	0.702 2	0.424 10	0.749 1	0.137 10	0.000 3	0.142 9	0.360 9	0.863 4	0.305 9	0.877 5	0.000 3	0.173 3	0.606 9	0.337 9	0.478 9	0.154 10	0.000 1	0.253 9	0.664 5	0.000 4	0.000 6	0.000 1	0.000 3	0.626 8	0.782 9	0.302 11	0.602 4	0.185 10	0.282 6	0.651 8	0.317 8	0.000 3	0.000 5	0.000 1	0.022 8	0.000 3	0.154 1	0.876 6	0.000 1	0.014 6	0.063 7	0.029 12	0.553 4	0.467 2	0.084 8	0.124 9	0.157 11	0.049 9	0.373 9	0.252 7	0.097 10	0.000 3	0.219 4	0.542 2	0.000 1	0.392 3	0.172 5	0.000 10	0.339 6	0.417 6	0.533 9	0.093 10	0.115 5	0.195 7	0.000 1	0.516 7	0.288 11	0.741 2	0.000 1	0.001 6	0.233 4	0.056 9	0.000 5	0.159 4	0.334 7	0.077 6	0.000 4	0.000 8	0.000 1	0.000 1	0.749 9	0.000 1	0.411 5	0.000 1	0.008 7	0.452 8	0.000 1	0.595 7	0.000 1	0.000 1	0.220 7	0.006 7	0.894 8	0.006 4	0.000 6	0.000 1	0.000 5	0.000 1	0.112 4	0.504 6	0.404 7	0.551 1	0.093 4	0.129 12	0.484 7	0.381 12	0.000 8	0.000 1	0.396 10	0.000 2	0.000 2	0.620 2	0.402 12	0.000 2	0.000 1	0.000 1	0.142 8	0.000 4	0.000 2	0.512 4	0.000 1

CSC-Pretrain	0.249 12	0.455 12	0.171 11	0.079 12	0.766 12	0.659 10	0.930 12	0.494 9	0.542 12	0.700 12	0.314 12	0.215 12	0.430 12	0.121 1	0.697 12	0.441 11	0.683 11	0.235 9	0.609 12	0.895 11	0.476 12	0.816 11	0.770 12	0.186 9	0.634 4	0.216 12	0.734 6	0.340 11	0.471 11	0.307 11	0.293 12	0.591 12	0.542 10	0.076 6	0.205 11	0.464 10	0.000 1	0.484 12	0.832 12	0.766 7	0.052 11	0.000 7	0.413 11	0.059 11	0.418 11	0.222 11	0.318 12	0.609 10	0.206 10	0.112 6	0.743 9	0.625 9	0.076 7	0.579 11	0.548 8	0.590 11	0.371 11	0.552 12	0.081 11	0.003 2	0.142 9	0.201 12	0.638 12	0.233 11	0.686 12	0.000 3	0.142 7	0.444 12	0.375 8	0.247 12	0.198 9	0.000 1	0.128 12	0.454 12	0.019 2	0.097 1	0.000 1	0.000 3	0.553 9	0.557 11	0.373 8	0.545 11	0.164 11	0.014 12	0.547 11	0.174 10	0.000 3	0.002 3	0.000 1	0.037 3	0.000 3	0.063 8	0.664 11	0.000 1	0.000 7	0.130 2	0.170 9	0.152 12	0.335 7	0.079 9	0.110 10	0.175 9	0.098 6	0.175 12	0.166 10	0.045 12	0.207 1	0.014 9	0.465 4	0.000 1	0.001 12	0.001 12	0.046 7	0.299 10	0.327 11	0.537 8	0.033 11	0.012 12	0.186 8	0.000 1	0.205 11	0.377 9	0.463 11	0.000 1	0.058 3	0.000 7	0.055 10	0.041 1	0.000 7	0.105 11	0.000 7	0.000 4	0.000 8	0.000 1	0.000 1	0.398 10	0.000 1	0.308 12	0.000 1	0.000 8	0.319 10	0.000 1	0.543 9	0.000 1	0.000 1	0.062 10	0.004 8	0.862 11	0.000 5	0.000 6	0.000 1	0.000 5	0.000 1	0.123 3	0.316 11	0.225 10	0.250 10	0.094 3	0.180 5	0.332 11	0.441 6	0.000 8	0.000 1	0.310 12	0.000 2	0.000 2	0.000 7	0.592 9	0.000 2	0.000 1	0.000 1	0.203 2	0.000 4	0.000 2	0.000 6	0.000 1
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
CeCo	0.340 4	0.551 6	0.247 8	0.181 3	0.784 8	0.661 9	0.939 8	0.564 3	0.624 8	0.721 7	0.484 3	0.429 2	0.575 2	0.027 6	0.774 7	0.503 9	0.753 3	0.242 8	0.656 8	0.945 5	0.534 5	0.865 5	0.860 6	0.177 12	0.616 5	0.400 2	0.818 2	0.579 1	0.615 6	0.367 9	0.408 5	0.726 10	0.633 2	0.162 1	0.360 4	0.619 2	0.000 1	0.828 4	0.873 7	0.924 2	0.109 8	0.083 3	0.564 3	0.057 12	0.475 8	0.266 6	0.781 1	0.767 6	0.257 5	0.100 8	0.825 6	0.663 7	0.048 11	0.620 9	0.551 7	0.595 10	0.532 5	0.692 6	0.246 4	0.000 3	0.213 4	0.615 1	0.861 5	0.376 5	0.900 3	0.000 3	0.102 11	0.660 5	0.321 10	0.547 4	0.226 8	0.000 1	0.311 8	0.742 2	0.011 3	0.006 5	0.000 1	0.000 3	0.546 10	0.824 7	0.345 9	0.665 1	0.450 4	0.435 1	0.683 3	0.411 5	0.338 1	0.000 5	0.000 1	0.030 6	0.000 3	0.068 6	0.892 5	0.000 1	0.063 3	0.000 8	0.257 8	0.304 10	0.387 3	0.079 9	0.228 3	0.190 7	0.000 11	0.586 1	0.347 2	0.133 5	0.000 3	0.037 8	0.377 8	0.000 1	0.384 4	0.006 11	0.003 8	0.421 2	0.410 8	0.643 4	0.171 4	0.121 4	0.142 9	0.000 1	0.510 8	0.447 7	0.474 9	0.000 1	0.000 7	0.286 3	0.083 8	0.000 5	0.000 7	0.603 1	0.096 4	0.063 3	0.000 8	0.000 1	0.000 1	0.898 3	0.000 1	0.429 4	0.000 1	0.400 1	0.550 3	0.000 1	0.633 4	0.000 1	0.000 1	0.377 3	0.000 10	0.916 3	0.000 5	0.000 6	0.000 1	0.000 5	0.000 1	0.102 9	0.499 7	0.296 9	0.463 3	0.089 5	0.304 1	0.740 2	0.401 11	0.010 3	0.000 1	0.560 2	0.000 2	0.000 2	0.709 1	0.652 7	0.000 2	0.000 1	0.000 1	0.143 7	0.000 4	0.000 2	0.609 2	0.000 1
Zhisheng Zhong, Jiequan Cui, Yibo Yang, Xiaoyang Wu, Xiaojuan Qi, Xiangyu Zhang, Jiaya Jia: Understanding Imbalanced Semantic Segmentation Through Neural Collapse. CVPR 2023
Minkowski 34D	0.253 11	0.463 11	0.154 12	0.102 11	0.771 11	0.650 11	0.932 10	0.483 11	0.571 11	0.710 10	0.331 11	0.250 11	0.492 9	0.044 4	0.703 11	0.419 12	0.606 12	0.227 11	0.621 11	0.865 12	0.531 6	0.771 12	0.813 9	0.291 6	0.484 10	0.242 11	0.612 12	0.282 12	0.440 12	0.351 10	0.299 10	0.622 11	0.593 8	0.027 9	0.293 8	0.310 12	0.000 1	0.757 9	0.858 10	0.737 11	0.150 5	0.164 1	0.368 12	0.084 5	0.381 12	0.142 12	0.357 10	0.720 7	0.214 9	0.092 11	0.724 11	0.596 12	0.056 10	0.655 5	0.525 9	0.581 12	0.352 12	0.594 11	0.056 12	0.000 3	0.014 12	0.224 11	0.772 10	0.205 12	0.720 11	0.000 3	0.159 5	0.531 11	0.163 12	0.294 11	0.136 12	0.000 1	0.169 11	0.589 10	0.000 4	0.000 6	0.000 1	0.002 1	0.663 5	0.466 12	0.265 12	0.582 7	0.337 8	0.016 11	0.559 10	0.084 12	0.000 3	0.000 5	0.000 1	0.036 4	0.000 3	0.125 3	0.670 10	0.000 1	0.102 1	0.071 6	0.164 10	0.406 6	0.386 4	0.046 11	0.068 12	0.159 10	0.117 3	0.284 11	0.111 11	0.094 11	0.000 3	0.000 12	0.197 12	0.000 1	0.044 10	0.013 10	0.002 9	0.228 12	0.307 12	0.588 7	0.025 12	0.545 3	0.134 10	0.000 1	0.655 3	0.302 10	0.282 12	0.000 1	0.060 2	0.000 7	0.035 12	0.000 5	0.000 7	0.097 12	0.000 7	0.000 4	0.005 7	0.000 1	0.000 1	0.096 12	0.000 1	0.334 11	0.000 1	0.000 8	0.274 11	0.000 1	0.513 11	0.000 1	0.000 1	0.280 6	0.194 4	0.897 7	0.000 5	0.000 6	0.000 1	0.000 5	0.000 1	0.108 7	0.279 12	0.189 11	0.141 12	0.059 11	0.272 2	0.307 12	0.445 5	0.003 6	0.000 1	0.353 11	0.000 2	0.026 1	0.000 7	0.581 10	0.001 1	0.000 1	0.000 1	0.093 12	0.002 3	0.000 2	0.000 6	0.000 1
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
PTv3 ScanNet200	0.393 1	0.592 1	0.330 1	0.216 1	0.851 1	0.687 4	0.971 1	0.586 1	0.755 1	0.752 5	0.505 1	0.404 4	0.575 2	0.000 10	0.848 1	0.616 2	0.761 1	0.349 1	0.738 1	0.978 1	0.546 4	0.860 6	0.926 1	0.346 2	0.654 3	0.384 4	0.828 1	0.523 3	0.699 1	0.583 3	0.387 6	0.822 1	0.688 1	0.118 4	0.474 1	0.603 4	0.000 1	0.832 3	0.903 1	0.753 8	0.140 6	0.000 7	0.650 2	0.109 2	0.520 1	0.457 1	0.497 7	0.871 3	0.281 2	0.192 2	0.887 2	0.748 1	0.168 1	0.727 3	0.733 1	0.740 1	0.644 1	0.714 3	0.190 8	0.000 3	0.256 2	0.449 6	0.914 1	0.514 1	0.759 10	0.337 1	0.172 4	0.692 4	0.617 1	0.636 1	0.325 4	0.000 1	0.641 1	0.782 1	0.000 4	0.065 2	0.000 1	0.000 3	0.842 1	0.903 1	0.661 2	0.662 2	0.612 1	0.405 2	0.731 1	0.566 1	0.000 3	0.000 5	0.000 1	0.017 10	0.301 1	0.088 4	0.941 1	0.000 1	0.077 2	0.000 8	0.717 3	0.790 1	0.310 9	0.026 12	0.264 2	0.349 1	0.220 2	0.397 8	0.366 1	0.115 8	0.000 3	0.337 1	0.463 5	0.000 1	0.531 2	0.218 1	0.593 1	0.455 1	0.469 1	0.708 2	0.210 1	0.592 2	0.108 11	0.000 1	0.728 1	0.682 2	0.671 4	0.000 1	0.000 7	0.407 1	0.136 2	0.022 2	0.575 1	0.436 4	0.259 2	0.428 1	0.048 3	0.000 1	0.000 1	0.879 5	0.000 1	0.480 2	0.000 1	0.133 5	0.597 1	0.000 1	0.690 1	0.000 1	0.000 1	0.009 11	0.000 10	0.921 2	0.000 5	0.151 2	0.000 1	0.000 5	0.000 1	0.109 6	0.494 9	0.622 2	0.394 6	0.073 10	0.141 7	0.798 1	0.528 3	0.026 1	0.000 1	0.551 3	0.000 2	0.000 2	0.134 5	0.717 5	0.000 2	0.000 1	0.000 1	0.188 3	0.000 4	0.000 2	0.791 1	0.000 1
Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, Hengshuang Zhao: Point Transformer V3: Simpler, Faster, Stronger. CVPR 2024 (Oral)
L3DETR-ScanNet_200	0.336 5	0.533 8	0.279 3	0.155 5	0.801 7	0.689 2	0.946 4	0.539 6	0.660 5	0.759 2	0.380 9	0.333 9	0.583 1	0.000 10	0.788 6	0.529 7	0.740 5	0.261 7	0.679 6	0.940 8	0.525 8	0.860 6	0.883 4	0.226 8	0.613 6	0.397 3	0.720 8	0.512 4	0.565 7	0.620 1	0.417 4	0.775 8	0.629 3	0.158 2	0.298 7	0.579 8	0.000 1	0.835 1	0.883 4	0.927 1	0.114 7	0.079 4	0.511 7	0.073 8	0.508 3	0.312 3	0.629 3	0.861 4	0.192 11	0.098 10	0.908 1	0.636 8	0.032 12	0.563 12	0.514 10	0.664 3	0.505 7	0.697 5	0.225 7	0.000 3	0.264 1	0.411 8	0.860 6	0.321 8	0.960 1	0.058 2	0.109 9	0.776 1	0.526 3	0.557 2	0.303 6	0.000 1	0.339 7	0.712 4	0.000 4	0.014 4	0.000 1	0.000 3	0.638 7	0.856 3	0.641 5	0.579 8	0.107 12	0.119 10	0.661 6	0.416 4	0.000 3	0.000 5	0.000 1	0.007 12	0.000 3	0.067 7	0.910 3	0.000 1	0.000 7	0.000 8	0.463 6	0.448 5	0.294 11	0.324 1	0.293 1	0.211 5	0.108 5	0.448 6	0.068 12	0.141 4	0.000 3	0.330 2	0.699 1	0.000 1	0.256 6	0.192 3	0.000 10	0.355 5	0.418 5	0.209 12	0.146 7	0.679 1	0.101 12	0.000 1	0.503 10	0.687 1	0.671 4	0.000 1	0.000 7	0.174 6	0.117 3	0.000 5	0.122 5	0.515 2	0.104 3	0.259 2	0.312 2	0.000 1	0.000 1	0.765 8	0.000 1	0.369 9	0.000 1	0.183 4	0.422 9	0.000 1	0.646 2	0.000 1	0.000 1	0.565 1	0.001 9	0.125 12	0.010 3	0.002 5	0.000 1	0.487 1	0.000 1	0.075 11	0.548 3	0.420 6	0.233 11	0.082 7	0.138 10	0.430 9	0.427 8	0.000 8	0.000 1	0.549 4	0.000 2	0.000 2	0.074 6	0.409 11	0.000 2	0.000 1	0.000 1	0.152 6	0.051 2	0.000 2	0.598 3	0.000 1
Yanmin Wu, Qiankun Gao, Renrui Zhang, Jian Zhang: Language-Assisted 3D Scene Understanding. arXiv23.12

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

Method	avg ap 25%	head ap 25%	common ap 25%	tail ap 25%	chair	table	door	couch	cabinet	shelf	desk	office chair	bed	pillow	sink	picture	window	toilet	bookshelf	monitor	curtain	book	coffee table	box	refrigerator	lamp	kitchen cabinet	towel	clothes	tv	nightstand	counter	dresser	stool	cushion	plant	ceiling	bathtub	end table	dining table	keyboard	bag	backpack	toilet paper	printer	tv stand	whiteboard	blanket	shower curtain	trash can	closet	stairs	microwave	stove	shoe	computer tower	bottle	bin	bench	board	washing machine	mirror	copier	basket	sofa chair	file cabinet	fan	laptop	shower	paper	person	paper towel dispenser	oven	blinds	rack	plate	blackboard	piano	suitcase	radiator	recycling bin	wardrobe	soap dispenser	telephone	bucket	clock	stand	light	laundry basket	pipe	clothes dryer	seat	speaker	column	bicycle	ladder	bathroom stall	shower wall	cup	jacket	storage bin	coffee maker	dishwasher	paper towel roll	machine	mat	windowsill	bar	toaster	bulletin board	ironing board	kitchen counter	doorframe	toilet paper dispenser	mini fridge	fire extinguisher	ball	hat	shower curtain rod	water cooler	paper cutter	tray	ledge	mouse	cart	storage container	scale	tissue box	light switch	power outlet	decoration	sign	projector	vacuum cleaner	candle	plunger	stuffed animal	headphones	dish rack	broom	range hood	water bottle	vent	shower floor	water pitcher	mailbox	bowl	paper bag	music stand	projector screen	laundry detergent	object	bathroom vanity	laundry hamper	bathroom stall door	ceiling light	trash bin	dumbbell	stair rail	tube	bathroom cabinet	closet rod	coffee kettle	structure	shower head	keyboard piano	case of water bottles	coat rack	storage organizer	folded chair	fire alarm	power strip	calendar	poster

TD3D Scannet200	0.379 2	0.603 2	0.306 2	0.190 2	0.885 1	0.755 1	0.800 2	0.958 1	0.390 2	0.260 2	0.866 2	0.232 1	0.979 2	0.523 3	0.869 3	0.559 5	0.689 2	1.000 1	0.795 1	0.905 2	0.748 1	0.173 5	0.825 1	0.173 2	0.970 1	0.457 1	0.615 2	0.456 2	0.200 1	0.621 4	0.906 2	0.553 1	0.517 1	0.510 1	0.220 2	0.715 1	0.706 2	1.000 1	0.113 2	0.792 1	0.717 2	0.073 2	0.635 2	0.557 1	0.638 1	1.000 1	0.205 5	0.146 3	1.000 1	0.769 5	0.186 2	1.000 1	0.710 5	0.778 1	0.415 1	0.834 4	0.226 2	0.021 2	0.590 2	0.356 2	0.817 1	0.477 5	1.000 1	0.000 1	0.635 1	0.843 2	0.427 1	0.270 4	0.125 2	0.000 2	0.102 3	1.000 1	0.125 2	0.000 2	0.000 1	0.000 2	0.000 3	0.125 4	0.370 3	0.622 5	0.221 1	0.196 2	0.836 1	0.288 2	0.000 2	0.093 2	0.020 2	0.294 2	0.000 1	0.075 2	0.667 1	0.038 1	0.111 1	0.250 4	0.000 4	0.526 2	0.495 3	0.908 1	0.111 3	0.259 1	0.003 3	0.667 2	0.045 5	0.000 2	0.000 1	0.400 1	0.274 3	0.000 1	0.274 2	0.226 2	0.000 1	0.520 2	0.302 5	0.731 2	0.103 3	0.458 1	0.500 1	0.000 1	1.000 1	0.472 1	0.792 3	0.000 1	0.088 2	0.061 2	0.250 1	0.009 2	0.250 2	0.333 3	0.181 2	0.396 2	0.051 2	0.012 1	0.000 1	0.458 4	0.000 1	0.424 5	0.000 1	0.101 2	0.390 5	0.000 1	0.833 2	0.000 1	0.000 1	0.857 2	0.222 3	1.000 1	0.000 1	0.003 2	0.000 1	0.000 2	0.000 1	0.102 2	0.275 5	0.400 2	0.735 2	0.061 3	0.433 3	0.533 3	0.625 1	0.000 2	0.000 1	0.259 4	0.000 1	0.000 1	0.000 2	0.500 2	0.000 1	0.000 2	1.000 1	0.600 1	0.000 2	0.250 1	0.000 2	0.000 1
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
Minkowski 34D Inst.	0.280 4	0.488 4	0.192 5	0.124 4	0.804 4	0.518 4	0.772 5	0.904 3	0.337 5	0.191 4	0.443 4	0.000 3	0.861 4	0.502 4	0.868 4	0.669 4	0.587 4	0.997 3	0.467 5	0.828 5	0.732 2	0.342 3	0.745 3	0.119 5	0.918 2	0.404 5	0.419 4	0.398 3	0.172 3	0.618 5	0.743 4	0.167 2	0.077 5	0.500 2	0.000 3	0.568 4	0.506 5	1.000 1	0.044 4	0.000 3	0.502 4	0.010 4	0.593 4	0.284 5	0.305 5	0.903 5	0.213 4	0.142 4	0.981 3	0.790 4	0.000 4	1.000 1	0.715 4	0.538 5	0.346 4	0.830 5	0.067 3	0.000 3	0.400 3	0.074 4	0.333 4	0.551 2	1.000 1	0.000 1	0.292 3	0.777 4	0.118 5	0.317 3	0.100 4	0.000 2	0.191 2	0.648 3	0.000 3	0.000 2	0.000 1	0.000 2	0.000 3	0.500 1	0.213 5	0.825 1	0.021 5	0.333 1	0.648 5	0.098 4	0.000 2	0.000 3	0.000 3	0.077 3	0.000 1	0.000 5	0.150 5	0.000 3	0.000 3	0.000 5	0.225 2	0.281 4	0.447 4	0.000 5	0.090 4	0.148 4	0.000 4	0.479 5	0.542 1	0.000 2	0.000 1	0.200 3	0.131 5	0.000 1	0.250 3	0.000 4	0.000 1	0.159 5	0.396 4	0.677 3	0.021 4	0.000 4	0.500 1	0.000 1	1.000 1	0.442 3	0.125 5	0.000 1	0.000 4	0.000 3	0.000 4	0.333 1	0.000 3	0.528 1	0.000 3	0.000 3	0.000 3	0.000 3	0.000 1	0.200 5	0.000 1	0.516 4	0.000 1	0.000 3	0.500 3	0.000 1	0.833 2	0.000 1	0.000 1	0.286 4	0.083 4	0.750 3	0.000 1	0.000 3	0.000 1	0.000 2	0.000 1	0.059 5	0.445 3	0.200 3	0.535 4	0.070 2	0.167 4	0.385 4	0.375 3	0.000 2	0.000 1	0.333 3	0.000 1	0.000 1	0.000 2	0.500 2	0.000 1	0.000 2	0.000 2	0.200 3	0.000 2	0.000 2	0.000 2	0.000 1
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
CSC-Pretrain Inst.	0.275 5	0.466 5	0.218 4	0.110 5	0.783 5	0.383 5	0.783 4	0.829 5	0.367 4	0.168 5	0.305 5	0.000 3	0.661 5	0.413 5	0.869 2	0.719 1	0.546 5	0.997 3	0.685 4	0.841 4	0.555 5	0.277 4	0.768 2	0.132 3	0.779 5	0.448 3	0.364 5	0.212 5	0.161 4	0.768 2	0.692 5	0.000 4	0.395 3	0.500 2	0.000 3	0.450 5	0.591 3	1.000 1	0.020 5	0.000 3	0.423 5	0.007 5	0.625 3	0.420 3	0.505 3	1.000 1	0.353 2	0.119 5	0.571 4	0.819 2	0.014 3	1.000 1	0.774 2	0.689 4	0.311 5	0.866 2	0.067 3	0.000 3	0.400 3	0.000 5	0.278 5	0.501 3	1.000 1	0.000 1	0.162 5	0.584 5	0.286 3	0.206 5	0.125 2	0.000 2	0.084 4	0.649 2	0.000 3	0.000 2	0.000 1	0.000 2	0.000 3	0.125 4	0.312 4	0.727 3	0.221 2	0.000 4	0.667 4	0.114 3	0.000 2	0.000 3	0.000 3	0.065 5	0.000 1	0.004 4	0.278 3	0.000 3	0.000 3	0.500 2	0.000 4	0.571 1	0.000 5	0.250 4	0.019 5	0.145 5	0.000 4	0.667 2	0.200 4	0.000 2	0.000 1	0.200 3	0.258 4	0.000 1	0.000 4	0.000 4	0.000 1	0.369 4	0.429 3	0.613 4	0.000 5	0.000 4	0.500 1	0.000 1	0.500 5	0.333 5	0.500 4	0.000 1	0.106 1	0.000 3	0.000 4	0.000 3	0.000 3	0.333 3	0.000 3	0.000 3	0.000 3	0.000 3	0.000 1	0.918 1	0.000 1	0.638 1	0.000 1	0.000 3	0.750 1	0.000 1	0.833 2	0.000 1	0.000 1	0.143 5	0.000 5	0.750 3	0.000 1	0.000 3	0.000 1	0.000 2	0.000 1	0.063 4	0.377 4	0.200 3	0.222 5	0.055 4	0.500 2	0.677 2	0.250 4	0.000 2	0.000 1	0.500 2	0.000 1	0.000 1	0.000 2	0.500 2	0.000 1	0.000 2	0.000 2	0.115 5	0.000 2	0.000 2	0.000 2	0.000 1
Ji Hou, Benjamin Graham, Matthias Nießner, Saining Xie: Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts. CVPR 2021
LGround Inst.	0.314 3	0.529 3	0.225 3	0.155 3	0.810 3	0.625 3	0.798 3	0.940 2	0.372 3	0.217 3	0.484 3	0.000 3	0.927 3	0.528 2	0.826 5	0.694 2	0.605 3	1.000 1	0.731 2	0.846 3	0.716 3	0.350 2	0.589 5	0.123 4	0.857 4	0.457 2	0.578 3	0.376 4	0.183 2	0.765 3	0.800 3	0.000 4	0.278 4	0.500 2	0.000 3	0.659 2	0.569 4	1.000 1	0.093 3	0.000 3	0.539 3	0.010 3	0.578 5	0.378 4	0.571 2	1.000 1	0.337 3	0.252 1	0.530 5	0.814 3	0.000 4	0.744 5	0.743 3	0.746 3	0.346 3	0.863 3	0.067 3	0.000 3	0.400 3	0.167 3	0.667 3	0.488 4	1.000 1	0.000 1	0.208 4	0.783 3	0.166 4	0.375 2	0.071 5	0.000 2	0.200 1	0.607 4	0.000 3	0.000 2	0.000 1	0.000 2	1.000 1	0.500 1	0.517 1	0.716 4	0.221 2	0.000 4	0.706 3	0.085 5	0.000 2	0.000 3	0.000 3	0.077 4	0.000 1	0.063 3	0.278 3	0.000 3	0.000 3	0.500 2	0.083 3	0.181 5	0.515 2	0.286 3	0.144 1	0.219 2	0.042 1	0.582 4	0.400 3	0.000 2	0.000 1	0.000 5	0.305 2	0.000 1	0.000 4	0.036 3	0.000 1	0.413 3	0.500 2	0.533 5	0.250 2	0.200 2	0.500 1	0.000 1	1.000 1	0.472 1	1.000 1	0.000 1	0.000 4	0.000 3	0.250 1	0.000 3	0.000 3	0.333 3	0.000 3	0.000 3	0.000 3	0.000 3	0.000 1	0.600 3	0.000 1	0.594 2	0.000 1	0.000 3	0.500 3	0.000 1	0.647 5	0.000 1	0.000 1	0.429 3	0.333 2	0.500 5	0.000 1	0.000 3	0.000 1	0.000 2	0.000 1	0.069 3	0.696 1	0.050 5	0.556 3	0.031 5	0.042 5	0.750 1	0.250 4	0.000 2	0.000 1	0.630 1	0.000 1	0.000 1	0.000 2	0.500 2	0.000 1	0.000 2	0.000 2	0.400 2	0.000 2	0.000 2	0.000 2	0.000 1
David Rozenberszki, Or Litany, Angela Dai: Language-Grounded Indoor 3D Semantic Segmentation in the Wild.
Mask3D Scannet200	0.445 1	0.653 1	0.392 1	0.254 1	0.844 2	0.746 2	0.818 1	0.888 4	0.556 1	0.262 1	0.890 1	0.025 2	1.000 1	0.608 1	0.930 1	0.694 3	0.721 1	0.930 5	0.686 3	0.966 1	0.615 4	0.440 1	0.725 4	0.201 1	0.890 3	0.414 4	0.827 1	0.552 1	0.158 5	0.806 1	0.924 1	0.042 3	0.512 2	0.412 5	0.226 1	0.604 3	0.830 1	1.000 1	0.125 1	0.792 1	0.815 1	0.097 1	0.648 1	0.551 2	0.354 4	1.000 1	0.630 1	0.241 2	1.000 1	0.853 1	0.204 1	0.974 4	0.841 1	0.778 1	0.358 2	0.927 1	0.300 1	0.045 1	0.640 1	0.363 1	0.745 2	0.710 1	1.000 1	0.000 1	0.330 2	0.943 1	0.315 2	0.600 1	1.000 1	0.027 1	0.080 5	0.556 5	0.500 1	0.409 1	0.000 1	0.194 1	1.000 1	0.500 1	0.493 2	0.761 2	0.053 4	0.042 3	0.780 2	0.454 1	0.009 1	0.333 1	0.050 1	0.321 1	0.000 1	0.084 1	0.552 2	0.008 2	0.027 2	0.750 1	0.500 1	0.442 3	0.657 1	0.765 2	0.120 2	0.183 3	0.021 2	1.000 1	0.510 2	0.016 1	0.000 1	0.400 1	0.619 1	0.000 1	0.396 1	0.290 1	0.000 1	0.741 1	0.699 1	1.000 1	0.260 1	0.017 3	0.125 5	0.000 1	0.792 4	0.399 4	1.000 1	0.000 1	0.049 3	0.265 1	0.063 3	0.000 3	1.000 1	0.335 2	0.381 1	0.500 1	0.250 1	0.004 2	0.000 1	0.727 2	0.000 1	0.538 3	0.000 1	0.188 1	0.677 2	0.000 1	0.930 1	0.000 1	0.000 1	0.966 1	0.391 1	0.908 2	0.000 1	0.028 1	0.000 1	1.000 1	0.000 1	0.152 1	0.451 2	0.458 1	0.971 1	0.573 1	0.606 1	0.167 5	0.625 1	0.004 1	0.000 1	0.058 5	0.000 1	0.000 1	1.000 1	1.000 1	0.000 1	0.056 1	0.000 2	0.200 3	0.309 1	0.000 2	1.000 1	0.000 1
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023

ScanNet Benchmark

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

Method	avg iou	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	floor	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	wall	window

PTv3 ScanNet	0.794 1	0.941 3	0.813 19	0.851 7	0.782 6	0.890 2	0.597 1	0.916 2	0.696 8	0.713 3	0.979 1	0.635 1	0.384 2	0.793 2	0.907 8	0.821 4	0.790 31	0.696 11	0.967 3	0.903 1	0.805 1
Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, Hengshuang Zhao: Point Transformer V3: Simpler, Faster, Stronger. CVPR 2024 (Oral)
PonderV2	0.785 2	0.978 1	0.800 27	0.833 23	0.788 4	0.853 16	0.545 16	0.910 5	0.713 1	0.705 4	0.979 1	0.596 7	0.390 1	0.769 12	0.832 41	0.821 4	0.792 30	0.730 1	0.975 1	0.897 4	0.785 4
Haoyi Zhu, Honghui Yang, Xiaoyang Wu, Di Huang, Sha Zhang, Xianglong He, Tong He, Hengshuang Zhao, Chunhua Shen, Yu Qiao, Wanli Ouyang: PonderV2: Pave the Way for 3D Foundataion Model with A Universal Pre-training Paradigm.
Mix3D	0.781 3	0.964 2	0.855 1	0.843 16	0.781 7	0.858 12	0.575 6	0.831 33	0.685 14	0.714 2	0.979 1	0.594 8	0.310 26	0.801 1	0.892 16	0.841 2	0.819 4	0.723 4	0.940 13	0.887 6	0.725 24
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
Swin3D	0.779 4	0.861 20	0.818 14	0.836 20	0.790 3	0.875 4	0.576 5	0.905 6	0.704 5	0.739 1	0.969 10	0.611 2	0.349 10	0.756 22	0.958 1	0.702 45	0.805 14	0.708 7	0.916 32	0.898 3	0.801 2

TTT-KD	0.773 5	0.646 91	0.818 14	0.809 35	0.774 9	0.878 3	0.581 2	0.943 1	0.687 12	0.704 5	0.978 4	0.607 5	0.336 15	0.775 8	0.912 6	0.838 3	0.823 2	0.694 12	0.967 3	0.899 2	0.794 3
Lisa Weijler, Muhammad Jehanzeb Mirza, Leon Sick, Can Ekkazan, Pedro Hermosilla: TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Models.
ResLFE_HDS	0.772 6	0.939 4	0.824 6	0.854 6	0.771 10	0.840 30	0.564 10	0.900 8	0.686 13	0.677 12	0.961 16	0.537 31	0.348 11	0.769 12	0.903 10	0.785 10	0.815 6	0.676 22	0.939 14	0.880 11	0.772 8

OctFormer	0.766 7	0.925 7	0.808 23	0.849 9	0.786 5	0.846 26	0.566 9	0.876 15	0.690 10	0.674 14	0.960 17	0.576 17	0.226 67	0.753 24	0.904 9	0.777 12	0.815 6	0.722 5	0.923 27	0.877 13	0.776 7
Peng-Shuai Wang: OctFormer: Octree-based Transformers for 3D Point Clouds. SIGGRAPH 2023
PPT-SpUNet-Joint	0.766 7	0.932 5	0.794 33	0.829 25	0.751 22	0.854 14	0.540 20	0.903 7	0.630 34	0.672 15	0.963 14	0.565 21	0.357 8	0.788 3	0.900 12	0.737 25	0.802 15	0.685 17	0.950 7	0.887 6	0.780 5
Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao: Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training. CVPR 2024
OccuSeg+Semantic	0.764 9	0.758 58	0.796 31	0.839 18	0.746 25	0.907 1	0.562 11	0.850 25	0.680 16	0.672 15	0.978 4	0.610 3	0.335 17	0.777 6	0.819 45	0.847 1	0.830 1	0.691 14	0.972 2	0.885 8	0.727 22

CU-Hybrid Net	0.764 9	0.924 8	0.819 12	0.840 17	0.757 17	0.853 16	0.580 3	0.848 26	0.709 3	0.643 24	0.958 21	0.587 12	0.295 33	0.753 24	0.884 20	0.758 19	0.815 6	0.725 3	0.927 24	0.867 22	0.743 15

O-CNN	0.762 11	0.924 8	0.823 7	0.844 15	0.770 11	0.852 18	0.577 4	0.847 28	0.711 2	0.640 28	0.958 21	0.592 9	0.217 73	0.762 17	0.888 17	0.758 19	0.813 10	0.726 2	0.932 22	0.868 21	0.744 14
Peng-Shuai Wang, Yang Liu, Yu-Xiao Guo, Chun-Yu Sun, Xin Tong: O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis. SIGGRAPH 2017
DTC	0.757 12	0.843 26	0.820 10	0.847 12	0.791 2	0.862 10	0.511 33	0.870 17	0.707 4	0.652 20	0.954 35	0.604 6	0.279 44	0.760 18	0.942 2	0.734 26	0.766 44	0.701 10	0.884 54	0.874 19	0.736 16

OA-CNN-L_ScanNet20	0.756 13	0.783 44	0.826 5	0.858 4	0.776 8	0.837 33	0.548 15	0.896 11	0.649 26	0.675 13	0.962 15	0.586 13	0.335 17	0.771 11	0.802 49	0.770 15	0.787 33	0.691 14	0.936 17	0.880 11	0.761 10

PNE	0.755 14	0.786 42	0.835 4	0.834 22	0.758 15	0.849 21	0.570 8	0.836 32	0.648 27	0.668 17	0.978 4	0.581 16	0.367 6	0.683 35	0.856 29	0.804 6	0.801 19	0.678 19	0.961 5	0.889 5	0.716 29
P. Hermosilla: Point Neighborhood Embeddings.
ConDaFormer	0.755 14	0.927 6	0.822 8	0.836 20	0.801 1	0.849 21	0.516 30	0.864 22	0.651 25	0.680 11	0.958 21	0.584 15	0.282 41	0.759 20	0.855 31	0.728 28	0.802 15	0.678 19	0.880 59	0.873 20	0.756 12
Lunhao Duan, Shanshan Zhao, Nan Xue, Mingming Gong, Guisong Xia, Dacheng Tao: ConDaFormer : Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding. Neurips, 2023
DMF-Net	0.752 16	0.906 12	0.793 35	0.802 41	0.689 40	0.825 46	0.556 12	0.867 18	0.681 15	0.602 44	0.960 17	0.555 27	0.365 7	0.779 5	0.859 26	0.747 22	0.795 27	0.717 6	0.917 31	0.856 30	0.764 9
C.Yang, Y.Yan, W.Zhao, J.Ye, X.Yang, A.Hussain, B.Dong, K.Huang: Towards Deeper and Better Multi-view Feature Fusion for 3D Semantic Segmentation. ICONIP 2023
PointTransformerV2	0.752 16	0.742 66	0.809 22	0.872 1	0.758 15	0.860 11	0.552 13	0.891 13	0.610 41	0.687 6	0.960 17	0.559 25	0.304 29	0.766 15	0.926 4	0.767 16	0.797 23	0.644 33	0.942 11	0.876 16	0.722 26
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
PointConvFormer	0.749 18	0.793 40	0.790 36	0.807 37	0.750 24	0.856 13	0.524 26	0.881 14	0.588 53	0.642 27	0.977 8	0.591 10	0.274 47	0.781 4	0.929 3	0.804 6	0.796 24	0.642 34	0.947 9	0.885 8	0.715 30
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
BPNet	0.749 18	0.909 10	0.818 14	0.811 33	0.752 20	0.839 32	0.485 47	0.842 29	0.673 18	0.644 23	0.957 25	0.528 37	0.305 28	0.773 9	0.859 26	0.788 8	0.818 5	0.693 13	0.916 32	0.856 30	0.723 25
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
MSP	0.748 20	0.623 94	0.804 25	0.859 3	0.745 26	0.824 48	0.501 37	0.912 4	0.690 10	0.685 8	0.956 26	0.567 20	0.320 23	0.768 14	0.918 5	0.720 33	0.802 15	0.676 22	0.921 29	0.881 10	0.779 6

StratifiedFormer	0.747 21	0.901 13	0.803 26	0.845 14	0.757 17	0.846 26	0.512 32	0.825 36	0.696 8	0.645 22	0.956 26	0.576 17	0.262 58	0.744 29	0.861 25	0.742 23	0.770 42	0.705 8	0.899 44	0.860 27	0.734 17
Xin Lai, Jianhui Liu, Li Jiang, Liwei Wang, Hengshuang Zhao, Shu Liu, Xiaojuan Qi, Jiaya Jia: Stratified Transformer for 3D Point Cloud Segmentation. CVPR 2022
VMNet	0.746 22	0.870 18	0.838 2	0.858 4	0.729 31	0.850 20	0.501 37	0.874 16	0.587 54	0.658 19	0.956 26	0.564 22	0.299 31	0.765 16	0.900 12	0.716 36	0.812 11	0.631 39	0.939 14	0.858 28	0.709 31
Zeyu HU, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai: VMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation. ICCV 2021 (Oral)
Virtual MVFusion	0.746 22	0.771 52	0.819 12	0.848 11	0.702 37	0.865 9	0.397 85	0.899 9	0.699 6	0.664 18	0.948 55	0.588 11	0.330 19	0.746 28	0.851 35	0.764 17	0.796 24	0.704 9	0.935 18	0.866 23	0.728 20
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
DiffSeg3D2	0.745 24	0.725 75	0.814 18	0.837 19	0.751 22	0.831 40	0.514 31	0.896 11	0.674 17	0.684 9	0.960 17	0.564 22	0.303 30	0.773 9	0.820 44	0.713 39	0.798 22	0.690 16	0.923 27	0.875 17	0.757 11

Retro-FPN	0.744 25	0.842 27	0.800 27	0.767 55	0.740 27	0.836 35	0.541 18	0.914 3	0.672 19	0.626 32	0.958 21	0.552 28	0.272 49	0.777 6	0.886 19	0.696 46	0.801 19	0.674 25	0.941 12	0.858 28	0.717 27
Peng Xiang, Xin Wen, Yu-Shen Liu, Hui Zhang, Yi Fang, Zhizhong Han: Retrospective Feature Pyramid Network for Point Cloud Semantic Segmentation. ICCV 2023
EQ-Net	0.743 26	0.620 95	0.799 30	0.849 9	0.730 30	0.822 50	0.493 44	0.897 10	0.664 20	0.681 10	0.955 29	0.562 24	0.378 3	0.760 18	0.903 10	0.738 24	0.801 19	0.673 26	0.907 36	0.877 13	0.745 13
Zetong Yang, Li Jiang, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
LRPNet	0.742 27	0.816 35	0.806 24	0.807 37	0.752 20	0.828 44	0.575 6	0.839 31	0.699 6	0.637 29	0.954 35	0.520 40	0.320 23	0.755 23	0.834 39	0.760 18	0.772 39	0.676 22	0.915 34	0.862 25	0.717 27

SAT	0.742 27	0.860 21	0.765 49	0.819 28	0.769 12	0.848 23	0.533 22	0.829 34	0.663 21	0.631 31	0.955 29	0.586 13	0.274 47	0.753 24	0.896 14	0.729 27	0.760 50	0.666 28	0.921 29	0.855 32	0.733 18

LargeKernel3D	0.739 29	0.909 10	0.820 10	0.806 39	0.740 27	0.852 18	0.545 16	0.826 35	0.594 52	0.643 24	0.955 29	0.541 30	0.263 57	0.723 33	0.858 28	0.775 14	0.767 43	0.678 19	0.933 20	0.848 37	0.694 36
Yukang Chen, Jianhui Liu, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs. CVPR 2023
RPN	0.736 30	0.776 48	0.790 36	0.851 7	0.754 19	0.854 14	0.491 46	0.866 20	0.596 51	0.686 7	0.955 29	0.536 32	0.342 13	0.624 50	0.869 22	0.787 9	0.802 15	0.628 40	0.927 24	0.875 17	0.704 33

MinkowskiNet	0.736 30	0.859 22	0.818 14	0.832 24	0.709 35	0.840 30	0.521 28	0.853 24	0.660 23	0.643 24	0.951 45	0.544 29	0.286 39	0.731 31	0.893 15	0.675 55	0.772 39	0.683 18	0.874 66	0.852 35	0.727 22
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
IPCA	0.731 32	0.890 14	0.837 3	0.864 2	0.726 32	0.873 5	0.530 25	0.824 37	0.489 87	0.647 21	0.978 4	0.609 4	0.336 15	0.624 50	0.733 58	0.758 19	0.776 37	0.570 65	0.949 8	0.877 13	0.728 20

PointTransformer++	0.725 33	0.727 74	0.811 21	0.819 28	0.765 13	0.841 29	0.502 36	0.814 42	0.621 37	0.623 34	0.955 29	0.556 26	0.284 40	0.620 52	0.866 23	0.781 11	0.757 54	0.648 31	0.932 22	0.862 25	0.709 31

SparseConvNet	0.725 33	0.647 90	0.821 9	0.846 13	0.721 33	0.869 6	0.533 22	0.754 58	0.603 47	0.614 36	0.955 29	0.572 19	0.325 21	0.710 34	0.870 21	0.724 31	0.823 2	0.628 40	0.934 19	0.865 24	0.683 39

MatchingNet	0.724 35	0.812 37	0.812 20	0.810 34	0.735 29	0.834 37	0.495 43	0.860 23	0.572 61	0.602 44	0.954 35	0.512 42	0.280 43	0.757 21	0.845 37	0.725 30	0.780 35	0.606 50	0.937 16	0.851 36	0.700 35

INS-Conv-semantic	0.717 36	0.751 61	0.759 52	0.812 32	0.704 36	0.868 7	0.537 21	0.842 29	0.609 43	0.608 40	0.953 39	0.534 34	0.293 34	0.616 53	0.864 24	0.719 35	0.793 28	0.640 35	0.933 20	0.845 41	0.663 45

PointMetaBase	0.714 37	0.835 28	0.785 38	0.821 26	0.684 42	0.846 26	0.531 24	0.865 21	0.614 38	0.596 48	0.953 39	0.500 45	0.246 63	0.674 36	0.888 17	0.692 47	0.764 46	0.624 42	0.849 81	0.844 42	0.675 41

contrastBoundary	0.705 38	0.769 55	0.775 43	0.809 35	0.687 41	0.820 53	0.439 73	0.812 43	0.661 22	0.591 50	0.945 63	0.515 41	0.171 91	0.633 47	0.856 29	0.720 33	0.796 24	0.668 27	0.889 51	0.847 38	0.689 37
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
ClickSeg_Semantic	0.703 39	0.774 50	0.800 27	0.793 46	0.760 14	0.847 25	0.471 51	0.802 46	0.463 94	0.634 30	0.968 12	0.491 48	0.271 51	0.726 32	0.910 7	0.706 41	0.815 6	0.551 77	0.878 60	0.833 43	0.570 77

RFCR	0.702 40	0.889 15	0.745 63	0.813 31	0.672 45	0.818 57	0.493 44	0.815 41	0.623 35	0.610 38	0.947 57	0.470 57	0.249 62	0.594 56	0.848 36	0.705 42	0.779 36	0.646 32	0.892 49	0.823 49	0.611 60
Jingyu Gong, Jiachen Xu, Xin Tan, Haichuan Song, Yanyun Qu, Yuan Xie, Lizhuang Ma: Omni-Supervised Point Cloud Segmentation via Gradual Receptive Field Component Reasoning. CVPR2021
One Thing One Click	0.701 41	0.825 32	0.796 31	0.723 62	0.716 34	0.832 39	0.433 75	0.816 39	0.634 32	0.609 39	0.969 10	0.418 83	0.344 12	0.559 68	0.833 40	0.715 37	0.808 13	0.560 71	0.902 41	0.847 38	0.680 40

JSENet	0.699 42	0.881 17	0.762 50	0.821 26	0.667 46	0.800 69	0.522 27	0.792 49	0.613 39	0.607 41	0.935 83	0.492 47	0.205 78	0.576 61	0.853 33	0.691 49	0.758 52	0.652 30	0.872 69	0.828 46	0.649 49
Zeyu HU, Mingmin Zhen, Xuyang BAI, Hongbo Fu, Chiew-lan Tai: JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds. ECCV 2020
One-Thing-One-Click	0.693 43	0.743 65	0.794 33	0.655 85	0.684 42	0.822 50	0.497 42	0.719 68	0.622 36	0.617 35	0.977 8	0.447 70	0.339 14	0.750 27	0.664 74	0.703 44	0.790 31	0.596 55	0.946 10	0.855 32	0.647 50
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
PicassoNet-II	0.692 44	0.732 70	0.772 44	0.786 47	0.677 44	0.866 8	0.517 29	0.848 26	0.509 80	0.626 32	0.952 43	0.536 32	0.225 69	0.545 74	0.704 65	0.689 52	0.810 12	0.564 70	0.903 40	0.854 34	0.729 19
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
Feature_GeometricNet	0.690 45	0.884 16	0.754 56	0.795 44	0.647 53	0.818 57	0.422 77	0.802 46	0.612 40	0.604 42	0.945 63	0.462 60	0.189 86	0.563 67	0.853 33	0.726 29	0.765 45	0.632 38	0.904 38	0.821 52	0.606 64
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
FusionNet	0.688 46	0.704 80	0.741 67	0.754 59	0.656 48	0.829 42	0.501 37	0.741 63	0.609 43	0.548 58	0.950 49	0.522 39	0.371 4	0.633 47	0.756 53	0.715 37	0.771 41	0.623 43	0.861 77	0.814 55	0.658 46
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
Feature-Geometry Net	0.685 47	0.866 19	0.748 60	0.819 28	0.645 55	0.794 72	0.450 63	0.802 46	0.587 54	0.604 42	0.945 63	0.464 59	0.201 81	0.554 70	0.840 38	0.723 32	0.732 64	0.602 53	0.907 36	0.822 51	0.603 67

KP-FCNN	0.684 48	0.847 25	0.758 54	0.784 49	0.647 53	0.814 60	0.473 50	0.772 52	0.605 45	0.594 49	0.935 83	0.450 68	0.181 89	0.587 57	0.805 48	0.690 50	0.785 34	0.614 46	0.882 56	0.819 53	0.632 56
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
VACNN++	0.684 48	0.728 73	0.757 55	0.776 52	0.690 38	0.804 67	0.464 56	0.816 39	0.577 60	0.587 51	0.945 63	0.508 44	0.276 46	0.671 37	0.710 63	0.663 60	0.750 58	0.589 60	0.881 57	0.832 45	0.653 48

DGNet	0.684 48	0.712 79	0.784 39	0.782 51	0.658 47	0.835 36	0.499 41	0.823 38	0.641 29	0.597 47	0.950 49	0.487 50	0.281 42	0.575 62	0.619 78	0.647 68	0.764 46	0.620 45	0.871 72	0.846 40	0.688 38

PointContrast_LA_SEM	0.683 51	0.757 59	0.784 39	0.786 47	0.639 57	0.824 48	0.408 80	0.775 51	0.604 46	0.541 60	0.934 87	0.532 35	0.269 53	0.552 71	0.777 51	0.645 71	0.793 28	0.640 35	0.913 35	0.824 48	0.671 42

Superpoint Network	0.683 51	0.851 24	0.728 71	0.800 43	0.653 50	0.806 65	0.468 53	0.804 44	0.572 61	0.602 44	0.946 60	0.453 67	0.239 66	0.519 79	0.822 42	0.689 52	0.762 49	0.595 57	0.895 47	0.827 47	0.630 57

VI-PointConv	0.676 53	0.770 54	0.754 56	0.783 50	0.621 61	0.814 60	0.552 13	0.758 56	0.571 63	0.557 56	0.954 35	0.529 36	0.268 55	0.530 77	0.682 69	0.675 55	0.719 67	0.603 52	0.888 52	0.833 43	0.665 44
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
ROSMRF3D	0.673 54	0.789 41	0.748 60	0.763 57	0.635 59	0.814 60	0.407 82	0.747 60	0.581 58	0.573 53	0.950 49	0.484 51	0.271 51	0.607 54	0.754 54	0.649 65	0.774 38	0.596 55	0.883 55	0.823 49	0.606 64

SALANet	0.670 55	0.816 35	0.770 47	0.768 54	0.652 51	0.807 64	0.451 60	0.747 60	0.659 24	0.545 59	0.924 93	0.473 56	0.149 101	0.571 64	0.811 47	0.635 74	0.746 59	0.623 43	0.892 49	0.794 68	0.570 77

O3DSeg	0.668 56	0.822 33	0.771 46	0.496 105	0.651 52	0.833 38	0.541 18	0.761 55	0.555 69	0.611 37	0.966 13	0.489 49	0.370 5	0.388 99	0.580 81	0.776 13	0.751 56	0.570 65	0.956 6	0.817 54	0.646 51

PointConv	0.666 57	0.781 45	0.759 52	0.699 70	0.644 56	0.822 50	0.475 49	0.779 50	0.564 66	0.504 76	0.953 39	0.428 77	0.203 80	0.586 59	0.754 54	0.661 61	0.753 55	0.588 61	0.902 41	0.813 57	0.642 52
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointASNL	0.666 57	0.703 81	0.781 41	0.751 61	0.655 49	0.830 41	0.471 51	0.769 53	0.474 90	0.537 62	0.951 45	0.475 55	0.279 44	0.635 45	0.698 68	0.675 55	0.751 56	0.553 76	0.816 88	0.806 59	0.703 34
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PPCNN++	0.663 59	0.746 63	0.708 74	0.722 63	0.638 58	0.820 53	0.451 60	0.566 96	0.599 49	0.541 60	0.950 49	0.510 43	0.313 25	0.648 42	0.819 45	0.616 79	0.682 82	0.590 59	0.869 73	0.810 58	0.656 47
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
MVF-GNN	0.658 60	0.558 102	0.751 58	0.655 85	0.690 38	0.722 94	0.453 59	0.867 18	0.579 59	0.576 52	0.893 105	0.523 38	0.293 34	0.733 30	0.571 83	0.692 47	0.659 89	0.606 50	0.875 63	0.804 61	0.668 43

DCM-Net	0.658 60	0.778 46	0.702 77	0.806 39	0.619 62	0.813 63	0.468 53	0.693 76	0.494 83	0.524 68	0.941 75	0.449 69	0.298 32	0.510 81	0.821 43	0.675 55	0.727 66	0.568 68	0.826 86	0.803 62	0.637 54
Jonas Schult, Francis Engelmann, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
HPGCNN	0.656 62	0.698 83	0.743 65	0.650 87	0.564 79	0.820 53	0.505 35	0.758 56	0.631 33	0.479 80	0.945 63	0.480 53	0.226 67	0.572 63	0.774 52	0.690 50	0.735 62	0.614 46	0.853 80	0.776 83	0.597 70
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
SAFNet-seg	0.654 63	0.752 60	0.734 69	0.664 83	0.583 74	0.815 59	0.399 84	0.754 58	0.639 30	0.535 64	0.942 73	0.470 57	0.309 27	0.665 38	0.539 85	0.650 64	0.708 72	0.635 37	0.857 79	0.793 70	0.642 52
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Net	0.645 64	0.778 46	0.731 70	0.699 70	0.577 75	0.829 42	0.446 65	0.736 64	0.477 89	0.523 70	0.945 63	0.454 64	0.269 53	0.484 89	0.749 57	0.618 77	0.738 60	0.599 54	0.827 85	0.792 73	0.621 59

PointConv-SFPN	0.641 65	0.776 48	0.703 76	0.721 64	0.557 82	0.826 45	0.451 60	0.672 81	0.563 67	0.483 79	0.943 72	0.425 80	0.162 96	0.644 43	0.726 59	0.659 62	0.709 71	0.572 64	0.875 63	0.786 78	0.559 83

MVPNet	0.641 65	0.831 29	0.715 72	0.671 80	0.590 70	0.781 78	0.394 86	0.679 78	0.642 28	0.553 57	0.937 80	0.462 60	0.256 59	0.649 41	0.406 99	0.626 75	0.691 79	0.666 28	0.877 61	0.792 73	0.608 63
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
PointMRNet	0.640 67	0.717 78	0.701 78	0.692 73	0.576 76	0.801 68	0.467 55	0.716 69	0.563 67	0.459 86	0.953 39	0.429 76	0.169 93	0.581 60	0.854 32	0.605 80	0.710 69	0.550 78	0.894 48	0.793 70	0.575 75

FPConv	0.639 68	0.785 43	0.760 51	0.713 68	0.603 65	0.798 70	0.392 87	0.534 101	0.603 47	0.524 68	0.948 55	0.457 62	0.250 61	0.538 75	0.723 61	0.598 84	0.696 77	0.614 46	0.872 69	0.799 63	0.567 80
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
PD-Net	0.638 69	0.797 39	0.769 48	0.641 93	0.590 70	0.820 53	0.461 57	0.537 100	0.637 31	0.536 63	0.947 57	0.388 90	0.206 77	0.656 39	0.668 72	0.647 68	0.732 64	0.585 62	0.868 74	0.793 70	0.473 103

PointSPNet	0.637 70	0.734 69	0.692 85	0.714 67	0.576 76	0.797 71	0.446 65	0.743 62	0.598 50	0.437 91	0.942 73	0.403 86	0.150 100	0.626 49	0.800 50	0.649 65	0.697 76	0.557 74	0.846 82	0.777 82	0.563 81

SConv	0.636 71	0.830 30	0.697 81	0.752 60	0.572 78	0.780 80	0.445 67	0.716 69	0.529 73	0.530 65	0.951 45	0.446 71	0.170 92	0.507 84	0.666 73	0.636 73	0.682 82	0.541 84	0.886 53	0.799 63	0.594 71

Supervoxel-CNN	0.635 72	0.656 88	0.711 73	0.719 65	0.613 63	0.757 89	0.444 70	0.765 54	0.534 72	0.566 54	0.928 91	0.478 54	0.272 49	0.636 44	0.531 87	0.664 59	0.645 93	0.508 91	0.864 76	0.792 73	0.611 60

joint point-based	0.634 73	0.614 96	0.778 42	0.667 82	0.633 60	0.825 46	0.420 78	0.804 44	0.467 92	0.561 55	0.951 45	0.494 46	0.291 36	0.566 65	0.458 94	0.579 90	0.764 46	0.559 73	0.838 83	0.814 55	0.598 69
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
PointMTL	0.632 74	0.731 71	0.688 88	0.675 77	0.591 69	0.784 77	0.444 70	0.565 97	0.610 41	0.492 77	0.949 53	0.456 63	0.254 60	0.587 57	0.706 64	0.599 83	0.665 88	0.612 49	0.868 74	0.791 76	0.579 74

3DSM_DMMF	0.631 75	0.626 93	0.745 63	0.801 42	0.607 64	0.751 90	0.506 34	0.729 67	0.565 65	0.491 78	0.866 108	0.434 72	0.197 84	0.595 55	0.630 77	0.709 40	0.705 74	0.560 71	0.875 63	0.740 93	0.491 98

PointNet2-SFPN	0.631 75	0.771 52	0.692 85	0.672 78	0.524 87	0.837 33	0.440 72	0.706 74	0.538 71	0.446 88	0.944 69	0.421 82	0.219 72	0.552 71	0.751 56	0.591 86	0.737 61	0.543 83	0.901 43	0.768 85	0.557 84

APCF-Net	0.631 75	0.742 66	0.687 90	0.672 78	0.557 82	0.792 75	0.408 80	0.665 82	0.545 70	0.508 73	0.952 43	0.428 77	0.186 87	0.634 46	0.702 66	0.620 76	0.706 73	0.555 75	0.873 67	0.798 65	0.581 73
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
FusionAwareConv	0.630 78	0.604 98	0.741 67	0.766 56	0.590 70	0.747 91	0.501 37	0.734 65	0.503 82	0.527 66	0.919 97	0.454 64	0.323 22	0.550 73	0.420 98	0.678 54	0.688 80	0.544 81	0.896 46	0.795 67	0.627 58
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DenSeR	0.628 79	0.800 38	0.625 101	0.719 65	0.545 84	0.806 65	0.445 67	0.597 90	0.448 97	0.519 71	0.938 79	0.481 52	0.328 20	0.489 88	0.499 92	0.657 63	0.759 51	0.592 58	0.881 57	0.797 66	0.634 55

SegGroup_sem	0.627 80	0.818 34	0.747 62	0.701 69	0.602 66	0.764 86	0.385 91	0.629 87	0.490 85	0.508 73	0.931 90	0.409 85	0.201 81	0.564 66	0.725 60	0.618 77	0.692 78	0.539 85	0.873 67	0.794 68	0.548 87
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
SIConv	0.625 81	0.830 30	0.694 83	0.757 58	0.563 80	0.772 84	0.448 64	0.647 85	0.520 76	0.509 72	0.949 53	0.431 75	0.191 85	0.496 86	0.614 79	0.647 68	0.672 86	0.535 87	0.876 62	0.783 79	0.571 76

dtc_net	0.625 81	0.703 81	0.751 58	0.794 45	0.535 85	0.848 23	0.480 48	0.676 80	0.528 74	0.469 83	0.944 69	0.454 64	0.004 114	0.464 91	0.636 76	0.704 43	0.758 52	0.548 80	0.924 26	0.787 77	0.492 97

HPEIN	0.618 83	0.729 72	0.668 91	0.647 89	0.597 68	0.766 85	0.414 79	0.680 77	0.520 76	0.525 67	0.946 60	0.432 73	0.215 74	0.493 87	0.599 80	0.638 72	0.617 98	0.570 65	0.897 45	0.806 59	0.605 66
Li Jiang, Hengshuang Zhao, Shu Liu, Xiaoyong Shen, Chi-Wing Fu, Jiaya Jia: Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation. ICCV 2019
SPH3D-GCN	0.610 84	0.858 23	0.772 44	0.489 106	0.532 86	0.792 75	0.404 83	0.643 86	0.570 64	0.507 75	0.935 83	0.414 84	0.046 111	0.510 81	0.702 66	0.602 82	0.705 74	0.549 79	0.859 78	0.773 84	0.534 90
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
AttAN	0.609 85	0.760 57	0.667 92	0.649 88	0.521 88	0.793 73	0.457 58	0.648 84	0.528 74	0.434 93	0.947 57	0.401 87	0.153 99	0.454 92	0.721 62	0.648 67	0.717 68	0.536 86	0.904 38	0.765 86	0.485 99
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
wsss-transformer	0.600 86	0.634 92	0.743 65	0.697 72	0.601 67	0.781 78	0.437 74	0.585 93	0.493 84	0.446 88	0.933 88	0.394 88	0.011 113	0.654 40	0.661 75	0.603 81	0.733 63	0.526 88	0.832 84	0.761 88	0.480 100

LAP-D	0.594 87	0.720 76	0.692 85	0.637 94	0.456 98	0.773 83	0.391 89	0.730 66	0.587 54	0.445 90	0.940 77	0.381 91	0.288 37	0.434 95	0.453 96	0.591 86	0.649 91	0.581 63	0.777 92	0.749 92	0.610 62

DPC	0.592 88	0.720 76	0.700 79	0.602 98	0.480 94	0.762 88	0.380 92	0.713 72	0.585 57	0.437 91	0.940 77	0.369 93	0.288 37	0.434 95	0.509 91	0.590 88	0.639 96	0.567 69	0.772 94	0.755 90	0.592 72
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
CCRFNet	0.589 89	0.766 56	0.659 96	0.683 75	0.470 97	0.740 93	0.387 90	0.620 89	0.490 85	0.476 81	0.922 95	0.355 96	0.245 64	0.511 80	0.511 90	0.571 91	0.643 94	0.493 95	0.872 69	0.762 87	0.600 68

ROSMRF	0.580 90	0.772 51	0.707 75	0.681 76	0.563 80	0.764 86	0.362 94	0.515 102	0.465 93	0.465 85	0.936 82	0.427 79	0.207 76	0.438 93	0.577 82	0.536 94	0.675 85	0.486 96	0.723 100	0.779 80	0.524 93

SD-DETR	0.576 91	0.746 63	0.609 105	0.445 110	0.517 89	0.643 105	0.366 93	0.714 71	0.456 95	0.468 84	0.870 107	0.432 73	0.264 56	0.558 69	0.674 70	0.586 89	0.688 80	0.482 97	0.739 98	0.733 95	0.537 89

SQN_0.1%	0.569 92	0.676 85	0.696 82	0.657 84	0.497 90	0.779 81	0.424 76	0.548 98	0.515 78	0.376 98	0.902 104	0.422 81	0.357 8	0.379 100	0.456 95	0.596 85	0.659 89	0.544 81	0.685 103	0.665 106	0.556 85

TextureNet	0.566 93	0.672 87	0.664 93	0.671 80	0.494 92	0.719 95	0.445 67	0.678 79	0.411 103	0.396 96	0.935 83	0.356 95	0.225 69	0.412 97	0.535 86	0.565 92	0.636 97	0.464 99	0.794 91	0.680 103	0.568 79
Jingwei Huang, Haotian Zhang, Li Yi, Thomas Funkerhouser, Matthias Niessner, Leonidas Guibas: TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes. CVPR
DVVNet	0.562 94	0.648 89	0.700 79	0.770 53	0.586 73	0.687 99	0.333 98	0.650 83	0.514 79	0.475 82	0.906 101	0.359 94	0.223 71	0.340 102	0.442 97	0.422 105	0.668 87	0.501 92	0.708 101	0.779 80	0.534 90

Pointnet++ & Feature	0.557 95	0.735 68	0.661 95	0.686 74	0.491 93	0.744 92	0.392 87	0.539 99	0.451 96	0.375 99	0.946 60	0.376 92	0.205 78	0.403 98	0.356 102	0.553 93	0.643 94	0.497 93	0.824 87	0.756 89	0.515 94

GMLPs	0.538 96	0.495 107	0.693 84	0.647 89	0.471 96	0.793 73	0.300 101	0.477 103	0.505 81	0.358 101	0.903 103	0.327 99	0.081 108	0.472 90	0.529 88	0.448 103	0.710 69	0.509 89	0.746 96	0.737 94	0.554 86

PanopticFusion-label	0.529 97	0.491 108	0.688 88	0.604 97	0.386 103	0.632 106	0.225 111	0.705 75	0.434 100	0.293 107	0.815 109	0.348 97	0.241 65	0.499 85	0.669 71	0.507 96	0.649 91	0.442 105	0.796 90	0.602 110	0.561 82
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
subcloud_weak	0.516 98	0.676 85	0.591 108	0.609 95	0.442 99	0.774 82	0.335 97	0.597 90	0.422 102	0.357 102	0.932 89	0.341 98	0.094 107	0.298 104	0.528 89	0.473 101	0.676 84	0.495 94	0.602 109	0.721 98	0.349 110

Online SegFusion	0.515 99	0.607 97	0.644 99	0.579 100	0.434 100	0.630 107	0.353 95	0.628 88	0.440 98	0.410 94	0.762 113	0.307 101	0.167 94	0.520 78	0.403 100	0.516 95	0.565 101	0.447 103	0.678 104	0.701 100	0.514 95
Davide Menini, Suryansh Kumar, Martin R. Oswald, Erik Sandstroem, Cristian Sminchisescu, Luc van Gool: A Real-Time Learning Framework for Joint 3D Reconstruction and Semantic Segmentation. Robotics and Automation Letters Submission
3DMV, FTSDF	0.501 100	0.558 102	0.608 106	0.424 112	0.478 95	0.690 98	0.246 107	0.586 92	0.468 91	0.450 87	0.911 99	0.394 88	0.160 97	0.438 93	0.212 109	0.432 104	0.541 107	0.475 98	0.742 97	0.727 96	0.477 101

PCNN	0.498 101	0.559 101	0.644 99	0.560 102	0.420 102	0.711 97	0.229 109	0.414 104	0.436 99	0.352 103	0.941 75	0.324 100	0.155 98	0.238 109	0.387 101	0.493 97	0.529 108	0.509 89	0.813 89	0.751 91	0.504 96

Weakly-Openseg v3	0.489 102	0.749 62	0.664 93	0.646 91	0.496 91	0.559 111	0.122 114	0.577 94	0.257 114	0.364 100	0.805 110	0.198 112	0.096 106	0.510 81	0.496 93	0.361 109	0.563 102	0.359 112	0.777 92	0.644 107	0.532 92

3DMV	0.484 103	0.484 109	0.538 110	0.643 92	0.424 101	0.606 110	0.310 99	0.574 95	0.433 101	0.378 97	0.796 111	0.301 102	0.214 75	0.537 76	0.208 110	0.472 102	0.507 111	0.413 108	0.693 102	0.602 110	0.539 88
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PointCNN with RGB	0.458 104	0.577 100	0.611 104	0.356 114	0.321 111	0.715 96	0.299 103	0.376 108	0.328 110	0.319 105	0.944 69	0.285 104	0.164 95	0.216 112	0.229 107	0.484 99	0.545 106	0.456 101	0.755 95	0.709 99	0.475 102
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
FCPN	0.447 105	0.679 84	0.604 107	0.578 101	0.380 104	0.682 100	0.291 104	0.106 114	0.483 88	0.258 112	0.920 96	0.258 108	0.025 112	0.231 111	0.325 103	0.480 100	0.560 104	0.463 100	0.725 99	0.666 105	0.231 114
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
DGCNN_reproduce	0.446 106	0.474 110	0.623 102	0.463 108	0.366 106	0.651 103	0.310 99	0.389 107	0.349 108	0.330 104	0.937 80	0.271 106	0.126 103	0.285 105	0.224 108	0.350 111	0.577 100	0.445 104	0.625 107	0.723 97	0.394 106
Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon: Dynamic Graph CNN for Learning on Point Clouds. TOG 2019
PNET2	0.442 107	0.548 104	0.548 109	0.597 99	0.363 107	0.628 108	0.300 101	0.292 109	0.374 105	0.307 106	0.881 106	0.268 107	0.186 87	0.238 109	0.204 111	0.407 106	0.506 112	0.449 102	0.667 105	0.620 109	0.462 104

SurfaceConvPF	0.442 107	0.505 106	0.622 103	0.380 113	0.342 109	0.654 102	0.227 110	0.397 106	0.367 106	0.276 109	0.924 93	0.240 109	0.198 83	0.359 101	0.262 105	0.366 107	0.581 99	0.435 106	0.640 106	0.668 104	0.398 105
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
Tangent Convolutions	0.438 109	0.437 112	0.646 98	0.474 107	0.369 105	0.645 104	0.353 95	0.258 111	0.282 112	0.279 108	0.918 98	0.298 103	0.147 102	0.283 106	0.294 104	0.487 98	0.562 103	0.427 107	0.619 108	0.633 108	0.352 109
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
3DWSSS	0.425 110	0.525 105	0.647 97	0.522 103	0.324 110	0.488 114	0.077 115	0.712 73	0.353 107	0.401 95	0.636 115	0.281 105	0.176 90	0.340 102	0.565 84	0.175 115	0.551 105	0.398 109	0.370 115	0.602 110	0.361 108

SPLAT Net	0.393 111	0.472 111	0.511 111	0.606 96	0.311 112	0.656 101	0.245 108	0.405 105	0.328 110	0.197 113	0.927 92	0.227 111	0.000 116	0.001 116	0.249 106	0.271 114	0.510 109	0.383 111	0.593 110	0.699 101	0.267 112
Hang Su, Varun Jampani, Deqing Sun, Subhransu Maji, Evangelos Kalogerakis, Ming-Hsuan Yang, Jan Kautz: SPLATNet: Sparse Lattice Networks for Point Cloud Processing. CVPR 2018
ScanNet+FTSDF	0.383 112	0.297 114	0.491 112	0.432 111	0.358 108	0.612 109	0.274 105	0.116 113	0.411 103	0.265 110	0.904 102	0.229 110	0.079 109	0.250 107	0.185 112	0.320 112	0.510 109	0.385 110	0.548 111	0.597 113	0.394 106

PointNet++	0.339 113	0.584 99	0.478 113	0.458 109	0.256 114	0.360 115	0.250 106	0.247 112	0.278 113	0.261 111	0.677 114	0.183 113	0.117 104	0.212 113	0.145 114	0.364 108	0.346 115	0.232 115	0.548 111	0.523 114	0.252 113
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNet	0.308 114	0.353 113	0.290 115	0.278 115	0.166 115	0.553 112	0.169 113	0.286 110	0.147 115	0.148 115	0.908 100	0.182 114	0.064 110	0.023 115	0.018 116	0.354 110	0.363 113	0.345 113	0.546 113	0.685 102	0.278 111

ScanNet	0.306 115	0.203 115	0.366 114	0.501 104	0.311 112	0.524 113	0.211 112	0.002 116	0.342 109	0.189 114	0.786 112	0.145 115	0.102 105	0.245 108	0.152 113	0.318 113	0.348 114	0.300 114	0.460 114	0.437 115	0.182 115
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
ERROR	0.054 116	0.000 116	0.041 116	0.172 116	0.030 116	0.062 116	0.001 116	0.035 115	0.004 116	0.051 116	0.143 116	0.019 116	0.003 115	0.041 114	0.050 115	0.003 116	0.054 116	0.018 116	0.005 116	0.264 116	0.082 116

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

Method	avg ap 25%	bathtub	bed	bookshelf	cabinet	chair	counter	curtain	desk	door	otherfurniture	picture	refrigerator	shower curtain	sink	sofa	table	toilet	window

OneFormer3D	0.896 1	1.000 1	1.000 1	0.913 5	0.858 5	0.951 5	0.786 11	0.837 16	0.916 11	0.908 1	0.778 5	0.803 4	0.750 12	1.000 1	0.976 3	0.926 4	0.882 5	0.995 43	0.849 1
Maxim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: OneFormer3D: One Transformer for Unified Point Cloud Segmentation.
MG-Former	0.887 2	1.000 1	0.991 11	0.837 22	0.801 19	0.935 14	0.887 2	0.857 9	0.946 3	0.891 6	0.748 13	0.805 3	0.739 14	1.000 1	0.993 2	0.809 53	0.876 12	1.000 1	0.842 2

UniPerception	0.884 3	1.000 1	0.979 17	0.872 15	0.869 2	0.892 23	0.806 8	0.890 5	0.835 26	0.892 5	0.755 10	0.811 1	0.779 9	0.955 43	0.951 4	0.876 21	0.914 1	0.997 36	0.840 3

InsSSM	0.883 4	1.000 1	0.996 3	0.800 35	0.865 3	0.960 2	0.808 7	0.852 13	0.940 5	0.899 4	0.785 3	0.810 2	0.700 18	1.000 1	0.912 15	0.851 38	0.895 2	0.997 36	0.827 5

TST3D	0.879 5	1.000 1	0.994 6	0.921 4	0.807 18	0.939 11	0.771 12	0.887 6	0.923 9	0.862 13	0.722 18	0.768 11	0.756 11	1.000 1	0.910 25	0.904 6	0.836 22	0.999 35	0.824 7

SIM3D	0.878 6	1.000 1	0.972 21	0.863 17	0.817 16	0.952 4	0.821 5	0.783 27	0.890 15	0.902 3	0.735 16	0.797 5	0.799 8	1.000 1	0.931 12	0.893 12	0.853 18	1.000 1	0.792 13

EV3D	0.877 7	1.000 1	0.996 5	0.873 13	0.854 6	0.950 6	0.691 21	0.783 28	0.926 6	0.889 9	0.754 11	0.794 8	0.820 2	1.000 1	0.912 15	0.900 8	0.860 16	1.000 1	0.779 16

TD3D	0.875 8	1.000 1	0.976 20	0.877 11	0.783 25	0.970 1	0.889 1	0.828 17	0.945 4	0.803 18	0.713 20	0.720 20	0.709 16	1.000 1	0.936 10	0.934 3	0.873 13	1.000 1	0.791 14
Maksim Kolodiazhnyi, Anna Vorontsova, Anton Konushin, Danila Rukhovich: Top-Down Beats Bottom-Up in 3D Instance Segmentation. WACV 2024
Spherical Mask(CtoF)	0.875 8	1.000 1	0.991 12	0.873 13	0.850 7	0.946 8	0.691 21	0.752 32	0.926 6	0.889 8	0.759 8	0.794 7	0.820 2	1.000 1	0.912 15	0.900 8	0.878 9	1.000 1	0.769 18

Queryformer	0.874 10	1.000 1	0.978 19	0.809 33	0.876 1	0.936 13	0.702 18	0.716 37	0.920 10	0.875 12	0.766 6	0.772 10	0.818 5	1.000 1	0.995 1	0.916 5	0.892 3	1.000 1	0.767 19

SoftGroup++	0.874 10	1.000 1	0.972 22	0.947 1	0.839 10	0.898 22	0.556 36	0.913 2	0.881 18	0.756 20	0.828 2	0.748 15	0.821 1	1.000 1	0.937 9	0.937 1	0.887 4	1.000 1	0.821 8

Mask3D	0.870 12	1.000 1	0.985 14	0.782 42	0.818 15	0.938 12	0.760 13	0.749 33	0.923 8	0.877 11	0.760 7	0.785 9	0.820 2	1.000 1	0.912 15	0.864 32	0.878 9	0.983 49	0.825 6
Jonas Schult, Francis Engelmann, Alexander Hermans, Or Litany, Siyu Tang, Bastian Leibe: Mask3D for 3D Semantic Instance Segmentation. ICRA 2023
ExtMask3D	0.867 13	1.000 1	1.000 1	0.756 49	0.816 17	0.940 10	0.795 9	0.760 31	0.862 20	0.888 10	0.739 14	0.763 12	0.774 10	1.000 1	0.929 13	0.878 20	0.879 7	1.000 1	0.819 10

SoftGroup	0.865 14	1.000 1	0.969 23	0.860 18	0.860 4	0.913 18	0.558 33	0.899 3	0.911 12	0.760 19	0.828 1	0.736 17	0.802 7	0.981 40	0.919 14	0.875 22	0.877 11	1.000 1	0.820 9
Thang Vu, Kookhoi Kim, Tung M. Luu, Xuan Thanh Nguyen, Chang D. Yoo: SoftGroup for 3D Instance Segmentaiton on Point Clouds. CVPR 2022 [Oral]
MAFT	0.860 15	1.000 1	0.990 13	0.810 32	0.829 11	0.949 7	0.809 6	0.688 43	0.836 25	0.904 2	0.751 12	0.796 6	0.741 13	1.000 1	0.864 35	0.848 40	0.837 20	1.000 1	0.828 4

IPCA-Inst	0.851 16	1.000 1	0.968 24	0.884 10	0.842 9	0.862 35	0.693 20	0.812 22	0.888 17	0.677 32	0.783 4	0.698 21	0.807 6	1.000 1	0.911 22	0.865 31	0.865 15	1.000 1	0.757 22

SPFormer	0.851 16	1.000 1	0.994 7	0.806 34	0.774 27	0.942 9	0.637 25	0.849 14	0.859 22	0.889 7	0.720 19	0.730 18	0.665 23	1.000 1	0.911 22	0.868 30	0.873 14	1.000 1	0.796 12
Sun Jiahao, Qing Chunmei, Tan Junpeng, Xu Xiangmin: Superpoint Transformer for 3D Scene Instance Segmentation. AAAI 2023 [Oral]
Mask3D_evaluation	0.843 18	1.000 1	0.955 29	0.847 20	0.795 21	0.932 15	0.750 15	0.780 29	0.891 14	0.818 15	0.737 15	0.633 30	0.703 17	1.000 1	0.902 27	0.870 26	0.820 23	0.941 57	0.805 11

ISBNet	0.835 19	1.000 1	0.950 30	0.731 51	0.819 13	0.918 16	0.790 10	0.740 34	0.851 24	0.831 14	0.661 28	0.742 16	0.650 26	1.000 1	0.937 8	0.814 52	0.836 21	1.000 1	0.765 20
Tuan Duc Ngo, Binh-Son Hua, Khoi Nguyen: ISBNet: a 3D Point Cloud Instance Segmentation Network with Instance-aware Sampling and Box-aware Dynamic Convolution. CVPR 2023
SphereSeg	0.835 19	1.000 1	0.963 27	0.891 8	0.794 22	0.954 3	0.822 4	0.710 38	0.961 2	0.721 24	0.693 26	0.530 43	0.653 25	1.000 1	0.867 34	0.857 35	0.859 17	0.991 46	0.771 17

TopoSeg	0.832 21	1.000 1	0.981 16	0.933 2	0.819 14	0.826 44	0.524 42	0.841 15	0.811 30	0.681 31	0.759 9	0.687 22	0.727 15	0.981 40	0.911 22	0.883 16	0.853 19	1.000 1	0.756 23

GraphCut	0.832 21	1.000 1	0.922 44	0.724 53	0.798 20	0.902 21	0.701 19	0.856 11	0.859 21	0.715 25	0.706 21	0.748 14	0.640 37	1.000 1	0.934 11	0.862 33	0.880 6	1.000 1	0.729 25

PBNet	0.825 23	1.000 1	0.963 26	0.837 24	0.843 8	0.865 30	0.822 3	0.647 46	0.878 19	0.733 22	0.639 35	0.683 23	0.650 26	1.000 1	0.853 36	0.870 27	0.820 24	1.000 1	0.744 24
W.Zhao, Y.Yan, C.Yang, J.Ye,X.Yang,K.Huang: Divide and Conquer: 3D Instance Segmentation With Point-Wise Binarization. ICCV 2023
SSEC	0.820 24	1.000 1	0.983 15	0.924 3	0.826 12	0.817 47	0.415 51	0.899 4	0.793 34	0.673 33	0.731 17	0.636 28	0.653 24	1.000 1	0.939 7	0.804 55	0.878 8	1.000 1	0.780 15

DKNet	0.815 25	1.000 1	0.930 36	0.844 21	0.765 31	0.915 17	0.534 40	0.805 24	0.805 32	0.807 17	0.654 29	0.763 13	0.650 26	1.000 1	0.794 48	0.881 17	0.766 28	1.000 1	0.758 21
Yizheng Wu, Min Shi, Shuaiyuan Du, Hao Lu, Zhiguo Cao, Weicai Zhong: 3D Instances as 1D Kernels. ECCV 2022
RPGN	0.806 26	1.000 1	0.992 9	0.789 37	0.723 44	0.891 24	0.650 24	0.810 23	0.832 27	0.665 35	0.699 24	0.658 24	0.700 18	1.000 1	0.881 29	0.832 44	0.774 26	0.997 36	0.613 45
Shichao Dong, Guosheng Lin, Tzu-Yi Hung: Learning Regional Purity for Instance Segmentation on 3D Point Clouds. ECCV 2022
Box2Mask	0.803 27	1.000 1	0.962 28	0.874 12	0.707 48	0.887 27	0.686 23	0.598 51	0.961 1	0.715 26	0.694 25	0.469 48	0.700 18	1.000 1	0.912 15	0.902 7	0.753 33	0.997 36	0.637 39
Julian Chibane, Francis Engelmann, Tuan Anh Tran, Gerard Pons-Moll: Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes. ECCV 2022
HAIS	0.803 27	1.000 1	0.994 7	0.820 28	0.759 32	0.855 36	0.554 37	0.882 7	0.827 29	0.615 41	0.676 27	0.638 27	0.646 35	1.000 1	0.912 15	0.797 58	0.767 27	0.994 44	0.726 26
Shaoyu Chen, Jiemin Fang, Qian Zhang, Wenyu Liu, Xinggang Wang: Hierarchical Aggregation for 3D Instance Segmentation. ICCV 2021
Mask-Group	0.792 29	1.000 1	0.968 25	0.812 29	0.766 30	0.864 31	0.460 45	0.815 21	0.888 16	0.598 45	0.651 32	0.639 26	0.600 43	0.918 46	0.941 5	0.896 11	0.721 40	1.000 1	0.723 27
Min Zhong, Xinghao Chen, Xiaokang Chen, Gang Zeng, Yunhe Wang: MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation. ICME 2022
CSC-Pretrained	0.791 30	1.000 1	0.996 3	0.829 27	0.767 29	0.889 26	0.600 28	0.819 20	0.770 39	0.594 46	0.620 39	0.541 40	0.700 18	1.000 1	0.941 5	0.889 14	0.763 29	1.000 1	0.526 55

SSTNet	0.789 31	1.000 1	0.840 58	0.888 9	0.717 45	0.835 40	0.717 17	0.684 44	0.627 54	0.724 23	0.652 31	0.727 19	0.600 43	1.000 1	0.912 15	0.822 47	0.757 32	1.000 1	0.691 33
Zhihao Liang, Zhihao Li, Songcen Xu, Mingkui Tan, Kui Jia: Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks. ICCV2021
GICN	0.788 32	1.000 1	0.978 18	0.867 16	0.781 26	0.833 41	0.527 41	0.824 18	0.806 31	0.549 54	0.596 42	0.551 36	0.700 18	1.000 1	0.853 36	0.935 2	0.733 37	1.000 1	0.651 36

DENet	0.786 33	1.000 1	0.929 37	0.736 50	0.750 38	0.720 60	0.755 14	0.934 1	0.794 33	0.590 47	0.561 48	0.537 41	0.650 26	1.000 1	0.882 28	0.804 56	0.789 25	1.000 1	0.719 28

DANCENET	0.786 33	1.000 1	0.936 33	0.783 40	0.737 41	0.852 38	0.742 16	0.647 46	0.765 41	0.811 16	0.624 38	0.579 33	0.632 40	1.000 1	0.909 26	0.898 10	0.696 45	0.944 53	0.601 48

DualGroup	0.782 35	1.000 1	0.927 38	0.811 30	0.772 28	0.853 37	0.631 27	0.805 24	0.773 36	0.613 42	0.611 40	0.610 31	0.650 26	0.835 57	0.881 29	0.879 19	0.750 35	1.000 1	0.675 34

PointGroup	0.778 36	1.000 1	0.900 48	0.798 36	0.715 46	0.863 32	0.493 43	0.706 39	0.895 13	0.569 52	0.701 22	0.576 34	0.639 38	1.000 1	0.880 31	0.851 37	0.719 41	0.997 36	0.709 30
Li Jiang, Hengshuang Zhao, Shaoshuai Shi, Shu Liu, Chi-Wing Fu, Jiaya Jia: PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation. CVPR 2020 [oral]
PE	0.776 37	1.000 1	0.900 49	0.860 18	0.728 43	0.869 28	0.400 52	0.857 10	0.774 35	0.568 53	0.701 23	0.602 32	0.646 35	0.933 45	0.843 39	0.890 13	0.691 49	0.997 36	0.709 29
Biao Zhang, Peter Wonka: Point Cloud Instance Segmentation using Probabilistic Embeddings. CVPR 2021
AOIA	0.767 38	1.000 1	0.937 32	0.810 31	0.740 40	0.906 19	0.550 38	0.800 26	0.706 46	0.577 51	0.624 37	0.544 39	0.596 48	0.857 49	0.879 33	0.880 18	0.750 34	0.992 45	0.658 35

DD-UNet+Group	0.764 39	1.000 1	0.897 51	0.837 23	0.753 35	0.830 43	0.459 47	0.824 18	0.699 48	0.629 39	0.653 30	0.438 51	0.650 26	1.000 1	0.880 31	0.858 34	0.690 50	1.000 1	0.650 37
H. Liu, R. Liu, K. Yang, J. Zhang, K. Peng, R. Stiefelhagen: HIDA: Towards Holistic Indoor Understanding for the Visually Impaired via Semantic Instance Segmentation with a Wearable Solid-State LiDAR Sensor. ICCVW 2021
INS-Conv-instance	0.762 40	1.000 1	0.923 41	0.765 45	0.785 24	0.905 20	0.600 28	0.655 45	0.646 53	0.683 30	0.647 33	0.530 42	0.650 26	1.000 1	0.824 41	0.830 45	0.693 48	0.944 53	0.644 38

Dyco3D	0.761 41	1.000 1	0.935 34	0.893 7	0.752 37	0.863 33	0.600 28	0.588 52	0.742 43	0.641 37	0.633 36	0.546 38	0.550 50	0.857 49	0.789 50	0.853 36	0.762 30	0.987 47	0.699 31
Tong He; Chunhua Shen; Anton van den Hengel: DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution. CVPR2021
OccuSeg+instance	0.742 42	1.000 1	0.923 41	0.785 38	0.745 39	0.867 29	0.557 34	0.578 55	0.729 44	0.670 34	0.644 34	0.488 46	0.577 49	1.000 1	0.794 48	0.830 45	0.620 58	1.000 1	0.550 51
Lei Han, Tian Zheng, Lan Xu, Lu Fang: OccuSeg: Occupancy-aware 3D Instance Segmentation. CVPR2020
RWSeg	0.739 43	1.000 1	0.899 50	0.759 47	0.753 36	0.823 45	0.282 57	0.691 42	0.658 51	0.582 50	0.594 43	0.547 37	0.628 41	1.000 1	0.795 47	0.868 29	0.728 39	1.000 1	0.692 32

3D-MPA	0.737 44	1.000 1	0.933 35	0.785 38	0.794 23	0.831 42	0.279 59	0.588 52	0.695 49	0.616 40	0.559 49	0.556 35	0.650 26	1.000 1	0.809 45	0.875 23	0.696 46	1.000 1	0.608 47
Francis Engelmann, Martin Bokeloh, Alireza Fathi, Bastian Leibe, Matthias Nießner: 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation. CVPR 2020
MTML	0.731 45	1.000 1	0.992 9	0.779 44	0.609 57	0.746 55	0.308 56	0.867 8	0.601 57	0.607 43	0.539 52	0.519 44	0.550 50	1.000 1	0.824 41	0.869 28	0.729 38	1.000 1	0.616 43
Jean Lahoud, Bernard Ghanem, Marc Pollefeys, Martin R. Oswald: 3D Instance Segmentation via Multi-task Metric Learning. ICCV 2019 [oral]
OSIS	0.725 46	1.000 1	0.885 54	0.653 59	0.657 54	0.801 48	0.576 32	0.695 41	0.828 28	0.698 28	0.534 53	0.457 50	0.500 57	0.857 49	0.831 40	0.841 42	0.627 56	1.000 1	0.619 42

SSEN	0.724 47	1.000 1	0.926 39	0.781 43	0.661 52	0.845 39	0.596 31	0.529 58	0.764 42	0.653 36	0.489 59	0.461 49	0.500 57	0.859 48	0.765 51	0.872 25	0.761 31	1.000 1	0.577 49
Dongsu Zhang, Junha Chun, Sang Kyun Cha, Young Min Kim: Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning. Arxiv
NeuralBF	0.718 48	1.000 1	0.945 31	0.901 6	0.754 34	0.817 46	0.460 45	0.700 40	0.772 37	0.688 29	0.568 47	0.000 70	0.500 57	0.981 40	0.606 61	0.872 24	0.740 36	1.000 1	0.614 44
Weiwei Sun, Daniel Rebain, Renjie Liao, Vladimir Tankovich, Soroosh Yazdani, Kwang Moo Yi, Andrea Tagliasacchi: NeuralBF: Neural Bilateral Filtering for Top-down Instance Segmentation on Point Clouds. WACV 2023
Sparse R-CNN	0.714 49	1.000 1	0.926 40	0.694 54	0.699 50	0.890 25	0.636 26	0.516 59	0.693 50	0.743 21	0.588 44	0.369 55	0.601 42	0.594 63	0.800 46	0.886 15	0.676 51	0.986 48	0.546 52

SALoss-ResNet	0.695 50	1.000 1	0.855 56	0.579 64	0.589 59	0.735 58	0.484 44	0.588 52	0.856 23	0.634 38	0.571 46	0.298 56	0.500 57	1.000 1	0.824 41	0.818 48	0.702 44	0.935 60	0.545 53
Zhidong Liang, Ming Yang, Hao Li, Chunxiang Wang: 3D Instance Embedding Learning With a Structure-Aware Loss Function for Point Cloud Segmentation. IEEE Robotics and Automation Letters (IROS2020)
PanopticFusion-inst	0.693 51	1.000 1	0.852 57	0.655 58	0.616 56	0.788 50	0.334 54	0.763 30	0.771 38	0.457 64	0.555 50	0.652 25	0.518 54	0.857 49	0.765 51	0.732 64	0.631 54	0.944 53	0.577 50
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
Occipital-SCS	0.688 52	1.000 1	0.913 45	0.730 52	0.737 42	0.743 57	0.442 48	0.855 12	0.655 52	0.546 55	0.546 51	0.263 58	0.508 56	0.889 47	0.568 62	0.771 61	0.705 43	0.889 63	0.625 41

3D-BoNet	0.687 53	1.000 1	0.887 53	0.836 25	0.587 60	0.643 67	0.550 38	0.620 48	0.724 45	0.522 59	0.501 57	0.243 59	0.512 55	1.000 1	0.751 53	0.807 54	0.661 53	0.909 62	0.612 46
Bo Yang, Jianan Wang, Ronald Clark, Qingyong Hu, Sen Wang, Andrew Markham, Niki Trigoni: Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds. NeurIPS 2019 Spotlight
ClickSeg_Instance	0.685 54	1.000 1	0.818 60	0.600 62	0.715 47	0.795 49	0.557 34	0.533 57	0.591 59	0.601 44	0.519 55	0.429 53	0.638 39	0.938 44	0.706 56	0.817 50	0.624 57	0.944 53	0.502 57

PCJC	0.684 55	1.000 1	0.895 52	0.757 48	0.659 53	0.862 34	0.189 66	0.739 35	0.606 56	0.712 27	0.581 45	0.515 45	0.650 26	0.857 49	0.357 67	0.785 59	0.631 55	0.889 63	0.635 40

SPG_WSIS	0.678 56	1.000 1	0.880 55	0.836 25	0.701 49	0.727 59	0.273 61	0.607 50	0.706 47	0.541 57	0.515 56	0.174 62	0.600 43	0.857 49	0.716 55	0.846 41	0.711 42	1.000 1	0.506 56

One_Thing_One_Click	0.675 57	1.000 1	0.823 59	0.782 41	0.621 55	0.766 52	0.211 63	0.736 36	0.560 61	0.586 48	0.522 54	0.636 29	0.453 61	0.641 61	0.853 36	0.850 39	0.694 47	0.997 36	0.411 62
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
SegGroup_ins	0.637 58	1.000 1	0.923 43	0.593 63	0.561 61	0.746 56	0.143 68	0.504 60	0.766 40	0.485 62	0.442 60	0.372 54	0.530

Presenting the ScanNet200 Benchmark

ScanNet200 Benchmark

ScanNet200 3D Semantic Label Benchmark

ScanNet200 3D Semantic Instance Benchmark

ScanNet Benchmark

3D Semantic Label Benchmark

3D Semantic Instance Benchmark