Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation

Benchmark Model Rank Results
semi-supervised-semantic-segmentation-on-nuscenesLaserMix (Voxel)#1mIoU (0.5% Labels): 51.4mIoU (2% Labels): 63.9mIoU (5% Labels): 69.7
semi-supervised-semantic-segmentation-on-nuscenesPLE (Voxel)#2mIoU (0.5% Labels): 58mIoU (1% Labels): 62.9mIoU (2% Labels): 67.2…
semi-supervised-semantic-segmentation-on-semantickittiPLE (Voxel)#1mIoU (0.5% Labels): 52.2mIoU (1% Labels): 61.1mIoU (2% Labels): 62.9…
semi-supervised-semantic-segmentation-on-semantickittiLaserMix (Voxel)#2mIoU (0.5% Labels): 47.3mIoU (2% Labels): 59.2mIoU (5% Labels): 61.7
semi-supervised-semantic-segmentation-on-semantickittiPLE (CENet, Range view)#3mIoU (0.5% Labels): 46.2mIoU (1% Labels): 51.5mIoU (2% Labels): 54.3…