Home News Latest Papers Tutorials Datasets Wiki SOTA LLM Models GPU Leaderboard Events

English

Semi Supervised Semantic Segmentation On 24

Metrics

mIOU (1% Test set)

mIoU (1% Labels)

mIoU (10% Labels)

mIoU (20% Labels)

mIoU (50% Labels)

Results

Performance results of various models on this benchmark

Model Name	mIOU (1% Test set)	mIoU (1% Labels)	mIoU (10% Labels)	mIoU (20% Labels)	mIoU (50% Labels)	Paper Title	Repository
360° from a Single Camera: A Few-Shot Approach for LiDAR Segmentation (All)	57.7	59.5	62.4	64.2	66.1	360$^circ$ from a Single Camera: A Few-Shot Approach for LiDAR Segmentation	-
PLE (Voxel)	-	61.1	63.1	64.1	64.3	Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation
MeanTeacher (Range View)	-	37.5	53.1	56.1	57.4	Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
LaserMix (Voxel)	-	50.6	60.0	61.9	62.3	LaserMix for Semi-Supervised LiDAR Semantic Segmentation
PLE (CENet, Range view)	-	51.5	-	-	-	Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation
CutMix-Seg (Range View)	-	37.4	54.3	56.6	57.6	Semi-supervised semantic segmentation needs strong, varied perturbations
CPS (Range View)	-	36.5	52.3	56.3	57.4	Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
SAPCA (Cylinder3D)	-	50.9	64.0	-	64.9	What Can be Seen is What You Get: Structure Aware Point Cloud Augmentation	-
MeanTeacher (Voxel)	-	45.4	57.1	59.2	60.0	Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
CBST (Range View)	-	39.9	53.4	56.1	56.9	Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training
LaserMix (Voxel)	-	-	-	-	-	Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation
LaserMix (Range View)	-	43.4	58.8	59.4	61.4	LaserMix for Semi-Supervised LiDAR Semantic Segmentation

0 of 12 row(s) selected.