3D Reconstruction on SemanticKITTI

Metric: mIoU (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	mIoU▼	Augmentations	Paper	Date↕	Code
1	TALoS	37.9	No	TALoS: Enhancing Semantic Scene Completion via T...	2024-10-21	Code
2	SCPNet	36.7	No	SCPNet: Semantic Scene Completion on Point Cloud	2023-03-13	Code
3	S3CNet	29.5	No	S3CNet: A Sparse Semantic Scene Completion Netwo...	2020-12-16	-
4	JS3C-Net	23.8	No	Sparse Single Sweep LiDAR Point Cloud Segmentati...	2020-12-07	Code
5	SSA-SC	23.5	No	Semantic Segmentation-assisted Scene Completion ...	2021-09-23	Code
6	Local-DIFs	22.7	No	Semantic Scene Completion using Local Deep Impli...	2020-11-18	-
7	UDNet	19.5	No	-	-	-
8	TS3D+DNet+SATNet (Reported in SemanticKITTI dataset paper)	17.7	No	Two Stream 3D Semantic Scene Completion	2018-04-10	-
9	LMSCNet-SS	17.6	No	LMSCNet: Lightweight Multiscale 3D Semantic Comp...	2020-08-24	Code
10	ESSCNet	17.5	No	Efficient Semantic Scene Completion Network with...	2019-07-11	Code
11	HTCL-S	17.09	No	Hierarchical Temporal Context Learning for Camer...	2024-07-02	Code
12	LMSCNet	17	No	LMSCNet: Lightweight Multiscale 3D Semantic Comp...	2020-08-24	Code
13	CGFormer	16.63	No	Context and Geometry Aware Voxel Transformer for...	2024-05-22	Code
14	SSCNet (reported in LMSCNet)	16.1	No	Semantic Scene Completion from a Single Depth Im...	2016-11-28	Code
15	SSCNet-full (reported in LMSCNet)	16.1	No	Semantic Scene Completion from a Single Depth Im...	2016-11-28	Code
16	OccDepth(RGB input only)	15.9	No	OccDepth: A Depth-Aware Method for 3D Semantic S...	2023-02-27	Code
17	Symphonies (RGB input only)	15.04	No	Symphonize 3D Semantic Scene Completion with Con...	2023-06-27	Code
18	Symphonies	15.04	No	Symphonize 3D Semantic Scene Completion with Con...	2023-06-27	Code
19	SG-SSC	12.9	No	-	-	Code
20	NDC-Scene	12.58	No	NDC-Scene: Boost Monocular 3D Semantic Scene Com...	2023-09-26	Code
21	OccFormer	12.32	No	OccFormer: Dual-path Transformer for Vision-base...	2023-04-11	Code
22	VoxFormer	12.2	No	VoxFormer: Sparse Voxel Transformer for Camera-b...	2023-02-23	Code
23	MonoScene (RGB input only)	11.08	No	MonoScene: Monocular 3D Semantic Scene Completion	2021-12-01	Code
24	MonoScene	11.08	No	MonoScene: Monocular 3D Semantic Scene Completion	2021-12-01	Code
25	TS3D+DNet (Reported in SemanticKITTI dataset paper)	10.2	No	Two Stream 3D Semantic Scene Completion	2018-04-10	-
26	TS3D (Reported in SemanticKITTI dataset paper)	9.5	No	Two Stream 3D Semantic Scene Completion	2018-04-10	-
27	JS3C-Net (rgb input - reported in MonoScene paper)	8.97	No	Sparse Single Sweep LiDAR Point Cloud Segmentati...	2020-12-07	Code
28	LMSCNet (rgb input - reported in MonoScene paper)	7.09	No	LMSCNet: Lightweight Multiscale 3D Semantic Comp...	2020-08-24	Code
29	3DSketch (rgb input - reported in MonoScene paper)	6.23	No	3D Sketch-aware Semantic Scene Completion via Se...	2020-03-31	Code

#1TALoSSOTA
37.9
mIoU· 2024-10-21
TALoS: Enhancing Semantic Scene Completion via Test-time Adaptation on the Line of Sight Code
#2SCPNetSOTA
36.7
mIoU· 2023-03-13
SCPNet: Semantic Scene Completion on Point Cloud Code
#3S3CNetSOTA
29.5
mIoU· 2020-12-16
S3CNet: A Sparse Semantic Scene Completion Network for LiDAR Point Clouds
#4JS3C-NetSOTA
23.8
mIoU· 2020-12-07
Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion Code
#5SSA-SC
23.5
mIoU· 2021-09-23
Semantic Segmentation-assisted Scene Completion for LiDAR Point Clouds Code
#6Local-DIFsSOTA
22.7
mIoU· 2020-11-18
Semantic Scene Completion using Local Deep Implicit Functions on LiDAR Data
#7UDNet
19.5
mIoU
No paper
#8TS3D+DNet+SATNet (Reported in SemanticKITTI dataset paper)SOTA
17.7
mIoU· 2018-04-10
Two Stream 3D Semantic Scene Completion
#9LMSCNet-SS
17.6
mIoU· 2020-08-24
LMSCNet: Lightweight Multiscale 3D Semantic Completion Code
#10ESSCNet
17.5
mIoU· 2019-07-11
Efficient Semantic Scene Completion Network with Spatial Group Convolution Code
#11HTCL-S
17.09
mIoU· 2024-07-02
Hierarchical Temporal Context Learning for Camera-based Semantic Scene Completion Code
#12LMSCNet
17
mIoU· 2020-08-24
LMSCNet: Lightweight Multiscale 3D Semantic Completion Code
#13CGFormer
16.63
mIoU· 2024-05-22
Context and Geometry Aware Voxel Transformer for Semantic Scene Completion Code
#14SSCNet (reported in LMSCNet)SOTA
16.1
mIoU· 2016-11-28
Semantic Scene Completion from a Single Depth Image Code
#15SSCNet-full (reported in LMSCNet)
16.1
mIoU· 2016-11-28
Semantic Scene Completion from a Single Depth Image Code
#16OccDepth(RGB input only)
15.9
mIoU· 2023-02-27
OccDepth: A Depth-Aware Method for 3D Semantic Scene Completion Code
#17Symphonies (RGB input only)
15.04
mIoU· 2023-06-27
Symphonize 3D Semantic Scene Completion with Contextual Instance Queries Code
#18Symphonies
15.04
mIoU· 2023-06-27
Symphonize 3D Semantic Scene Completion with Contextual Instance Queries Code
#19SG-SSC
12.9
mIoU
No paperCode
#20NDC-Scene
12.58
mIoU· 2023-09-26
NDC-Scene: Boost Monocular 3D Semantic Scene Completion in Normalized Device Coordinates Space Code
#21OccFormer
12.32
mIoU· 2023-04-11
OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Prediction Code
#22VoxFormer
12.2
mIoU· 2023-02-23
VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completion Code
#23MonoScene (RGB input only)
11.08
mIoU· 2021-12-01
MonoScene: Monocular 3D Semantic Scene Completion Code
#24MonoScene
11.08
mIoU· 2021-12-01
MonoScene: Monocular 3D Semantic Scene Completion Code
#25TS3D+DNet (Reported in SemanticKITTI dataset paper)
10.2
mIoU· 2018-04-10
Two Stream 3D Semantic Scene Completion
#26TS3D (Reported in SemanticKITTI dataset paper)
9.5
mIoU· 2018-04-10
Two Stream 3D Semantic Scene Completion
#27JS3C-Net (rgb input - reported in MonoScene paper)
8.97
mIoU· 2020-12-07
Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion Code
#28LMSCNet (rgb input - reported in MonoScene paper)
7.09
mIoU· 2020-08-24
LMSCNet: Lightweight Multiscale 3D Semantic Completion Code
#293DSketch (rgb input - reported in MonoScene paper)
6.23
mIoU· 2020-03-31
3D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Prior Code