AI Paper Atlas

Research map / Benchmarks / SemanticKITTI

SemanticKITTI leaderboard

98 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on SemanticKITTI. Best mIoU (val): 93.52 (TerraSeg-B, CVPR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

SemanticKITTI · mIoU (val)

67 papers, each with its best reported number for its own method; higher is better.

57 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

SemanticKITTI · mIoU

59 papers, each with its best reported number for its own method; higher is better.

#mIoU ↑MethodPaperTopic
176.1RAPiD-Seg
Minkowski-UNet34
RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation
ECCV 2024 · Table 1
Segmentation and detection
275.2UniSeg
Minkowski-UNet34 + SalsaNext + ResNet-34
UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebase
ICCV 2023 · Table 2
Segmentation and detection
375.1M3Net
PTv2+
Multi-Space Alignments Towards Universal LiDAR Segmentation
CVPR 2024 · Table 4
3D detection and driving
474.8SphereFormer
SparseConv
Spherical Transformer for LiDAR-Based 3D Recognition
CVPR 2023 · SemanticKITTI test results tab
Segmentation and detection
573.8UGSP
PTv3
Uncertainty-Driven Gaussian Sphere Propagation for 3D Semantic Segmentation
ECCV 2026 · Table 2
Segmentation and detection
673.6ProtoTransferProtoTransfer: Cross-Modal Prototype Transfer for Point Cloud Segmentation
ICCV 2023 · Table 2
Segmentation and detection
771.3PointMetaBase-L RS + FastAdapter
PointMetaBase-L
Mitigating Geometric Degradation in Fast DownSampling via FastAdapter for Point Cloud Segmentation
ICCV 2025 · Table 2
Point cloud classification
871.2C3D_0.5× + PVD*
Cylinder3D
Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation
CVPR 2022 · Table: SemanticKITTI test resu
Segmentation and detection
970.8WaffleIron
WaffleIron-48-256
Using a Waffle Iron for Automotive Point Cloud Semantic Segmentation
ICCV 2023 · Table 1
Segmentation and detection
1070.8GP-S3NetGP-S3Net: Graph-Based Panoptic Sparse Semantic Segmentation Network
ICCV 2021 · Table 1
CNN features
49 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

SemanticKITTI · IoU (val)

17 papers, each with its best reported number for its own method; higher is better.

#IoU ↑MethodPaperTopic
149.53C3DFusion (full)
CGFormer
Towards Temporal Fusion Beyond the Field of View for Camera-based Semantic Scene Completion
AAAI 2026 · Table 3
3D generation
249.1ConSSC full model
ResNet50, temporal
Learning Spatial-Temporal Consistency for 3D Semantic Scene Completion
CVPR 2026 · Table 3
3D generation
347.91FoundationSSC (full)Unleashing Semantic and Geometric Priors for 3D Scene Completion
AAAI 2026 · Table 3
3D generation
445.99Full CGFormerContext and Geometry Aware Voxel Transformer for Semantic Scene Completion
NeurIPS 2024 · Table: Ablation study of the a
3D generation
545.95ScanSSCThree Cars Approaching within 100m! Enhancing Distant Geometry by Tri-Axis Voxel Scanning for Camera-based Semantic Scene Completion
CVPR 2025 · Table 5
3D generation
645.65Shape Segments Interactive
EfficientNetB7 + SAM ViT-B
SDFormer: Vision-based 3D Semantic Scene Completion via SAM-assisted Dual-channel Voxel Transformer
ICCV 2025 · Table 6
3D generation
745.01SGFormerSGFormer: Satellite-Ground Fusion for 3D Semantic Scene Completion
CVPR 2025 · Table 1
3D generation
844.9VoxFormer-B + ProtoOcc
VoxFormer-B
3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation
CVPR 2025 · Table 2
3D generation
944.88Bi-SSCBi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion
CVPR 2024 · Table 2
3D generation
1044.71OctreeOccOctreeOcc: Efficient and Multi-Granularity Occupancy Prediction Using Octree Queries
NeurIPS 2024 · Table: 3D Semantic Scene Compl
3D generation
7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

SemanticKITTI · IoU

11 papers, each with its best reported number for its own method; higher is better.

1 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.