Segmentation and detection: research map
567 papers on Segmentation and detection in Point clouds, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 6 approaches. The busiest year so far is 2023.
What it is
Labelling every point in a 3D scan with a class (semantic segmentation) and finding 3D objects as boxes, mostly for driving and indoor scenes.
Terms you will see on this page:
- Voxel
- a 3D grid cell; many detectors turn points into voxels.
- Semantic segmentation
- assigning a class label to every point.
- 3D bounding box
- a box with position, size and heading around an object.
Start reading with the most cited papers here:
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (CVPR 2018 · 4,733 citations)
- PointPillars: Fast Encoders for Object Detection From Point Clouds (CVPR 2019 · 4,587 citations)
Within Point clouds, its share shrank from 21.9% in 2023–24 to 17.9% in 2025–26 (100 → 91 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Segmentation and detection in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Scene completion and instance segmentation rising: 13% → 21% of the topic (13 → 19 papers)
- Point cloud semantic segmentation rising: 32% → 36% of the topic (32 → 33 papers)
- LiDAR voxel and pillar detectors falling: 20% → 16% of the topic (20 → 15 papers)
- Multimodal and open-vocabulary 3D detection falling: 10% → 7% of the topic (10 → 6 papers)
Papers from 2025–2026 that are already cited most:
- UniMamba: Unified Spatial-Channel Representation Learning with Group-Efficient Mamba for LiDAR-based 3D Object Detection (CVPR 2025 · 17 citations)
- SpatialLM: Training Large Language Models for Structured Indoor Modeling (NeurIPS 2025 · 8 citations)
- Geometry-Aware 3D Salient Object Detection Network (AAAI 2025 · 2 citations)
Approaches and key papers
Point cloud segmentation · 360 papers
Keywords: semantic segmentation, cloud semantic, instance
Approaches in this cluster:
- Point cloud semantic segmentation (186 papers)
Segment point cloud scenes with weak supervision and joint semantic-instance learning. - LiDAR segmentation for driving (95 papers)
Segment LiDAR scans using domain adaptation, image distillation and efficient grid representations. - Scene completion and instance segmentation (79 papers)
Model object relations and open-vocabulary contrast for 3D instance segmentation and scene completion.
Most cited and most cited since 2024:
- RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds (CVPR 2020 · 2,138 citations)
- 3D Semantic Parsing of Large-Scale Indoor Spaces (CVPR 2016 · 1,979 citations)
- OneFormer3D: One Transformer for Unified Point Cloud Segmentation (CVPR 2024 · 142 citations)
- OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentation (CVPR 2024 · 66 citations)
3D object detection · 207 papers
Keywords: detection, 3d object, detectors
Approaches in this cluster:
- LiDAR voxel and pillar detectors (88 papers)
Detect 3D objects in driving LiDAR with efficient voxel, pillar and domain-adaptive designs. - Single-stage point-voxel detection (76 papers)
Detect 3D objects directly from point clouds with proposal-based and hybrid voxel-point networks. - Multimodal and open-vocabulary 3D detection (43 papers)
Improve 3D detection with transformers, vision fusion, multi-domain training and open-vocabulary labels.
Most cited and most cited since 2024:
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (CVPR 2018 · 4,733 citations)
- PointPillars: Fast Encoders for Object Detection From Point Clouds (CVPR 2019 · 4,587 citations)
- Voxel Mamba: Group-Free State Space Models for Point Cloud based 3D Object Detection (NeurIPS 2024 · 35 citations)
- LION: Linear Group RNN for 3D Object Detection in Point Clouds (NeurIPS 2024 · 29 citations)
Related topics in Point clouds
- Point cloud classification (1,097)
- Registration and matching (272)
- Shape and surface completion (558)
