Research map / Point clouds / Segmentation and detection
UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebase
Youquan Liu, Runnan Chen, Xin Li, Lingdong Kong, Yuchen Yang, Zhaoyang Xia, Yeqi Bai, Xinge Zhu, Yuexin Ma, Yikang Li, Yu Qiao, Yuenan Hou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add RGB image fusion via LMA on top of voxel+point+range views (ablation) | UniSeg · SemanticKITTI val | mIoU: 69.7 -> 71.3 | Improved |
| replace early-fusion addition with LMA deformable cross-attention fusion (ablation) | UniSeg · SemanticKITTI val | mIoU: 70.1 -> 71.3 | Improved |
| use input concatenation fusion instead of addition (ablation) | early fusion · SemanticKITTI val | mIoU: 70.1 -> 69.4 | Hurt |
| use PointPainting or PointAugmenting instead of LMA (ablation) | UniSeg · SemanticKITTI val | mIoU: 70.1 -> 70.5 | No clear effect |
| add LVA learnable cross-view fusion instead of addition (ablation) | UniSeg · SemanticKITTI val | mIoU: 70.4 -> 71.3 | Improved |
| use LMA instead of addition under noisy calibration matrices (ablation) | UniSeg · SemanticKITTI val | mIoU: 68.5 -> 71.0 | Improved |
| use voxel view only versus range view only (ablation) | single-view baseline · SemanticKITTI val | mIoU: 55.8 -> 68.4 | Improved |
| prune 80% channels per layer (UniSeg 0.2x) (ablation) | UniSeg · SemanticKITTI val | mIoU: 71.3 -> 67.0 | Hurt |
| use UniSeg versus Panoptic-PHNet for nuScenes panoptic | UniSeg · nuScenes test (panoptic) | PQ: 81.5 -> 78.4 | Hurt |
| reimplement methods in OpenPCSeg codebase | MinkowskiNet · SemanticKITTI val | mIoU: 61.1 -> 68.8 | Improved |
Later papers that compared with this method
1 later accepted paper reports UniSeg as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation (CVPR 2024 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- 2DPASS — 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds (ECCV 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Panoptic-PolarNet: Proposal-Free LiDAR Point Cloud Panoptic Segmentation (CVPR 2021 · similarity 0.90)
- Uni-to-Multi Modal Knowledge Distillation for Bidirectional LiDAR-Camera Semantic Segmentation (TPAMI 2024 · similarity 0.90)
- RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentation (ICCV 2021 · similarity 0.90)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-Based Perception (TPAMI 2021 · similarity 0.90)
- RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation (ECCV 2024 · similarity 0.90)
- How Do Images Align and Complement LiDAR? Towards a Harmonized Multi-modal 3D Panoptic Segmentation (ICML 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
