The 3D semantic labeling task involves predicting a semantic labeling of a 3D scan mesh.

Evaluation and metrics

Our evaluation ranks all methods according to the PASCAL VOC intersection-over-union metric (IoU). IoU = TP/(TP+FP+FN), where TP, FP, and FN are the numbers of true positive, false positive, and false negative pixels, respectively. Predicted labels are evaluated per-vertex over the respective 3D scan mesh; for 3D approaches that operate on other representations like grids or points, the predicted labels should be mapped onto the mesh vertices (e.g., one such example for grid to mesh vertices is provided in the evaluation helpers).



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




Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysorted bysort bysort bysort bysort bysort by
Q2E0.741 10.984 10.821 20.757 40.739 10.868 20.600 10.849 10.595 60.659 10.971 20.490 20.299 20.689 40.822 30.749 10.788 40.641 10.935 20.860 10.699 2
ActiveST0.735 20.983 20.769 40.798 10.701 20.852 50.527 20.801 20.680 10.629 20.973 10.447 100.312 10.757 10.799 40.747 20.795 30.632 20.952 10.855 20.684 3
Gengxin Liu, Oliver van Kaick, Hui Huang, Ruizhen Hu: Active Self-Training for Weakly Supervised 3D Scene Semantic Segmentation.
WS3D_LA_Sempermissive0.689 40.879 30.753 60.798 10.648 80.816 90.421 100.796 30.604 50.603 30.945 100.457 90.204 90.559 100.851 20.724 30.760 70.630 30.903 70.821 50.603 8
Kangcheng Liu: WS3D: Weakly Supervised 3D Scene Segmentation with Region-Level Boundary Awareness and Instance Discrimination. European Conference on Computer Vision (ECCV), 2022
DE-3DLearner LA0.704 30.774 70.766 50.764 30.687 40.832 70.413 110.790 40.639 20.599 40.952 40.478 60.222 80.746 20.859 10.678 40.806 20.607 60.915 50.847 30.703 1
Ping-Chung Yu, Cheng Sun, Min Sun: Data Efficient 3D Learner via Knowledge Transferred from 2D Model. ECCV 2022
GaIA0.682 60.731 110.846 10.713 80.657 60.869 10.475 40.705 90.452 130.569 50.951 50.563 10.290 30.544 110.799 40.677 50.810 10.618 40.900 80.821 50.642 5
Min Seok Lee*, Seok Woo Yang*, and Sung Won Han: GaIA: Graphical Information gain based Attention Network for Weakly Supervised 3D Point Cloud Semantic Segmentation. WACV 2023
CSC_LA_SEM0.644 100.761 80.707 120.703 100.642 100.813 100.436 80.659 110.502 90.516 110.945 100.487 30.238 70.538 120.678 90.659 60.739 120.568 100.915 50.811 80.566 12
One-Thing-One-Click0.670 80.734 100.815 30.661 130.644 90.841 60.509 30.741 70.479 120.548 70.968 30.461 80.251 50.664 60.754 70.656 70.744 100.541 110.917 40.844 40.625 6
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
VIBUSpermissive0.684 50.848 40.752 70.708 90.691 30.861 30.474 50.770 50.611 40.538 90.951 50.478 60.275 40.676 50.671 110.649 80.788 40.610 50.869 90.808 100.657 4
Beiwen Tian,Liyi Luo,Hao Zhao,Guyue Zhou: VIBUS: Data-efficient 3D Scene Parsing with VIewpoint Bottleneck and Uncertainty-Spectrum Modeling. ISPRS Journal of Photogrammetry and Remote Sensing
PointContrast_LA_SEM0.636 110.694 120.738 80.731 50.653 70.817 80.467 60.651 120.517 80.522 100.946 80.479 50.198 110.575 90.526 130.649 80.747 80.569 90.845 110.803 110.600 9
LE0.680 70.744 90.731 90.727 60.664 50.859 40.427 90.759 60.562 70.562 60.948 70.480 40.245 60.735 30.765 60.648 100.786 60.591 70.931 30.817 70.624 7
Viewpoint_BN_LA_AIR0.650 90.778 60.731 90.688 110.617 110.812 110.446 70.739 80.618 30.540 80.945 100.415 110.204 90.623 70.676 100.594 110.744 100.576 80.868 100.811 80.582 10
Liyi Luo, Beiwen Tian, Hao Zhao, Guyue Zhou: Pointly-supervised 3D Scene Parsing with Viewpoint Bottleneck.
SQN_LA0.576 130.674 130.670 130.722 70.454 130.790 130.342 130.622 130.487 100.427 130.933 130.357 130.157 130.452 130.721 80.492 120.696 130.487 130.790 130.748 130.507 13
Scratch_LA_SEM0.621 120.802 50.715 110.687 120.570 120.800 120.386 120.703 100.486 110.514 120.946 80.390 120.181 120.620 80.670 120.487 130.746 90.539 120.804 120.798 120.580 11