Semi Supervised Semantic Segmentation On 23

Metriken

mIoU (1% Labels)

mIoU (10% Labels)

mIoU (20% Labels)

mIoU (50% Labels)

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

					Paper Title
LaserMix (Voxel)	44.2	53.7	55.1	56.8	LaserMix for Semi-Supervised LiDAR Semantic Segmentation
MeanTeacher (Voxel)	41.0	50.1	52.8	53.9	Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
Sup.-only (Voxel)	39.2	48.0	52.1	53.8	Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
LaserMix (Range View)	38.3	54.4	55.6	58.7	LaserMix for Semi-Supervised LiDAR Semantic Segmentation
CutMix-Seg (Range View)	36.7	50.7	52.9	54.3	Semi-supervised semantic segmentation needs strong, varied perturbations
CBST (Range View)	35.7	50.7	52.7	54.6	Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training
MeanTeacher (Range View)	34.2	49.8	51.6	53.3	Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
CPS (Range View)	33.7	50.0	52.8	54.6	Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
Sup.-only (Range View)	33.1	47.7	49.9	52.5	FIDNet: LiDAR Point Cloud Semantic Segmentation with Fully Interpolation Decoding

0 of 9 row(s) selected.