10-shot image generation on SemanticKITTI

Metric: mIoU (1% Labels) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	mIoU (1% Labels)▼	Extra Data	Paper	Date↕	Code
1	PLE (Voxel)	61.1	No	Learning from Spatio-temporal Correlation for Se...	2024-10-09	Code
2	360° from a Single Camera: A Few-Shot Approach for LiDAR Segmentation (All)	59.5	Yes	360$^\circ$ from a Single Camera: A Few-Shot App...	2023-09-12	-
3	PLE (CENet, Range view)	51.5	No	Learning from Spatio-temporal Correlation for Se...	2024-10-09	Code
4	SAPCA (Cylinder3D)	50.9	No	What Can be Seen is What You Get: Structure Awar...	2022-06-20	-
5	LaserMix (Voxel)	50.6	No	LaserMix for Semi-Supervised LiDAR Semantic Segm...	2022-06-30	Code
6	MeanTeacher (Voxel)	45.4	No	Mean teachers are better role models: Weight-ave...	2017-03-06	Code
7	LaserMix (Range View)	43.4	No	LaserMix for Semi-Supervised LiDAR Semantic Segm...	2022-06-30	Code
8	CBST (Range View)	39.9	No	-	-	Code
9	MeanTeacher (Range View)	37.5	No	Mean teachers are better role models: Weight-ave...	2017-03-06	Code
10	CutMix-Seg (Range View)	37.4	No	Semi-supervised semantic segmentation needs stro...	2019-06-05	Code
11	CPS (Range View)	36.5	No	Semi-Supervised Semantic Segmentation with Cross...	2021-06-02	Code

#1PLE (Voxel)SOTA
61.1
mIoU (1% Labels)· 2024-10-09
Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation Code
#2360° from a Single Camera: A Few-Shot Approach for LiDAR Segmentation (All)SOTA
59.5
mIoU (1% Labels)· Extra Data· 2023-09-12
360$^\circ$ from a Single Camera: A Few-Shot Approach for LiDAR Segmentation
#3PLE (CENet, Range view)
51.5
mIoU (1% Labels)· 2024-10-09
Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation Code
#4SAPCA (Cylinder3D)SOTA
50.9
mIoU (1% Labels)· 2022-06-20
What Can be Seen is What You Get: Structure Aware Point Cloud Augmentation
#5LaserMix (Voxel)
50.6
mIoU (1% Labels)· 2022-06-30
LaserMix for Semi-Supervised LiDAR Semantic Segmentation Code
#6MeanTeacher (Voxel)SOTA
45.4
mIoU (1% Labels)· 2017-03-06
Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results Code
#7LaserMix (Range View)
43.4
mIoU (1% Labels)· 2022-06-30
LaserMix for Semi-Supervised LiDAR Semantic Segmentation Code
#8CBST (Range View)
39.9
mIoU (1% Labels)
No paperCode
#9MeanTeacher (Range View)
37.5
mIoU (1% Labels)· 2017-03-06
Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results Code
#10CutMix-Seg (Range View)
37.4
mIoU (1% Labels)· 2019-06-05
Semi-supervised semantic segmentation needs strong, varied perturbations Code
#11CPS (Range View)
36.5
mIoU (1% Labels)· 2021-06-02
Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision Code