Research map / Point clouds / Segmentation and detection
LiDAR segmentation for driving
Segment LiDAR scans using domain adaptation, image distillation and efficient grid representations.
47 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2020–2026) · an approach in Segmentation and detection, keyword cluster “semantic segmentation · cloud semantic · instance”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Quiet None of the tracked benchmarks has a report from 2025–26.
18 papers in 2023–24 and 14 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 18.0% to 15.4%.
No longer reported by papers here: SemanticKITTI test (mIoU, last in 2024); SemanticKITTI validation (mIoU, last in 2024); nuScenes validation (mIoU, last in 2024).
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2020 · PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020 · 597 citations)
- 2021 · Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · 599 citations)
- 2022 · Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation (CVPR 2022 · 198 citations)
- 2023 · RangeViT: Towards Vision Transformers for 3D Semantic Segmentation in Autonomous Driving (CVPR 2023 · 138 citations)
- 2024 · NeRF-LiDAR: Generating Realistic LiDAR Point Clouds with Neural Radiance Fields (AAAI 2024 · 51 citations)
- 2025 · LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenes (CVPR 2025 · 9 citations)
- 2026 · Adaptive Augmentation-Aware Latent Learning for Robust LiDAR Semantic Segmentation (ICLR 2026 · 0 citations)
Earliest papers here
- PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020 · 597 citations)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · 599 citations)
- Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion (AAAI 2021 · 276 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
SemanticKITTI test · mIoU (higher is better)
5 papers, 2020–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2020 | 54.3 | PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020 · 597 citations) |
| 2021 | 67.8 | Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · 599 citations) |
| 2022 | 71.2 | Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation (CVPR 2022 · 198 citations) |
SemanticKITTI validation · mIoU (higher is better)
4 papers, 2021–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2021 | 67.5 | Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion (AAAI 2021 · 276 citations) |
nuScenes validation · mIoU (higher is better)
4 papers, 2021–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2021 | 76.1 | Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · 599 citations) |
What has been tried
39 papers here report 495 measured changes (a component added, removed or swapped, compared with a baseline): helped 79%, no clear effect 4%, hurt 16%, mixed 1%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add additional vehicles through object placement | 3D vehicle detection · Average Precision | 25.7 -> 28.4 | Helped | NeRF-LiDAR: Generating Realistic LiDAR Point Clouds with Neural Radiance Fields (AAAI 2024 · 51 citations) |
| Add Region Caption Prompts | nuScenes 1% labels · mIoU | 38.8 -> 43.4 | Helped | VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding (AAAI 2024 · 24 citations) |
| Add Flexible Step Aggregation | SemanticKITTI val · mIoU | 68.9 -> 71.3 | Helped | TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation (CVPR 2024 · 18 citations) |
| remove semantic information | QEOxford · position error | 0.95m -> 1.79m | Hurt | LiSA: LiDAR Localization with Semantic Awareness (CVPR 2024 · 14 citations) |
| Use point and range experts alongside the voxel expert in CML | nuScenes LP · mIoU | 45.35 -> 46.02 | Helped | LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenes (CVPR 2025 · 9 citations) |
| add HTSF and LTSF rather than omit HEU | SemanticKITTI val, 10% labels · mIoU | 59.2 -> 65.7 | Helped | HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving (CVPR 2025 · 8 citations) |
| Remove reflectance from the input | SemanticKITTI→SemanticSTF · mIoU | 24.4 -> 37.3 | Helped | Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather (CVPR 2025 · 5 citations) |
| remove both ConEmb and ConAtt | SynLiDAR→SemKITTI · mIoU | 20.6 vs 23.3 full | Helped | Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation (CVPR 2024 · 5 citations) |
| Add Single Pattern LiDAR augmentation | Waymo, nuScenes, and SemanticKITTI, 1% labels · mIoU | 40.17 -> 43.44; 32.64 -> 34.88; 37.63 -> 39.09 | Helped | PSA-SSL: Pose and Size-aware Self-Supervised Learning on LiDAR Point Clouds (CVPR 2025 · 4 citations) |
| Add TMP to the annotation-free baseline | nuScenes annotation-free segmentation · mIoU | 44.16 -> 47.42 | Helped | 3D Annotation-Free Learning by Distilling 2D Open-Vocabulary Segmentation Models for Autonomous Driving (AAAI 2025 · 2 citations) |
| Combine image-based and LiDAR-based caption generation instead of using image captions only | nuScenes val · TPSS | 8.78 -> 9.65 | Helped | 3D-AVS: LiDAR-based 3D Auto-Vocabulary Segmentation (CVPR 2025 · 2 citations) |
| add decorrelation | SemanticKITTI-C · average mIoU | 24.25 -> 30.17 | Helped | D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation (CVPR 2025 · 2 citations) |
What reviewers will expect
Datasets and baselines named in the 8 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- nuScenes · 5 of 8 papers
- SemanticKITTI · 5 of 8 papers
- SemanticKITTI-C · 3 of 8 papers
- Synth4D · 2 of 8 papers
- nuScenes-C · 2 of 8 papers
- SemanticSTF · 2 of 8 papers
- SynLiDAR · 2 of 8 papers
Baselines
- CLIP2Scene · 2 of 8 papers
- OpenScene · 2 of 8 papers
- SLidR · 2 of 8 papers
- SEAL · 2 of 8 papers
- PolarMix · 2 of 8 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 32 papers; from OpenAlex and the papers' affiliations).
- Shanghai AI Laboratory · 3 papers
- Inceptio · 2 papers
- Beijing Institute of Technology · 2 papers
- Peking University · 2 papers
- Nanyang Technological University · 2 papers
- National University of Singapore · 2 papers
- Zhejiang University · 2 papers
- Shanghai Jiao Tong University · 2 papers
Use it
Download BibTeX of 38 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Point cloud semantic segmentation (108 papers)
- Scene completion and instance segmentation (44 papers)
Elsewhere in the topic
- LiDAR voxel and pillar detectors (56 papers)
- Single-stage point-voxel detection (45 papers)
- Multimodal and open-vocabulary 3D detection (24 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).