Research map / Point clouds / Segmentation and detection
Online Segmentation of LiDAR Sequences: Dataset and Algorithm
Romain Loiseau, Mathieu Aubry, Loïc Landrieu
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 the spatio-temporal transformer at the lowest resolution (instead of a 1x1x1 convolution) (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU: 58.6 -> 66.7 | Improved |
| add the spatio-temporal transformer at the lowest resolution (instead of a 1x1x1 convolution) (ablation) | Helix4D · SemanticKITTI val (online, 1/5 frame slices) | mIoU: 60.2 -> 66.8 | Improved |
| extend attention over past slices (temporal receptive field) rather than the current slice only (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU: 62.9 -> 66.7 | Improved |
| extend attention over past slices (temporal receptive field) rather than the current slice only (ablation) | Helix4D · SemanticKITTI val (online, 1/5 frame slices) | mIoU: 62.6 -> 66.8 | Improved |
| add relative spatio-temporal positional encoding to attention compatibilities (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU: 64.3 -> 66.7 | Improved |
| add relative spatio-temporal positional encoding to attention compatibilities (ablation) | Helix4D · SemanticKITTI val (online, 1/5 frame slices) | mIoU: 64.1 -> 66.8 | Improved |
| add separate queries to the simplified transformer (keys used as queries in the default) (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU: 66.7 -> 65.2 | Hurt |
| use Cylinder3D asymmetric convolutions instead of simple 3x3x3 convolutions (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU and inference time: 66.7 -> 66.6 mIoU; 45 -> 56 ms | No clear effect |
| shrink the model to 306k parameters with maxpool and narrower features (Helix4D Tiny) (ablation) | Helix4D · SemanticKITTI val (full frame) | mIoU: 66.7 -> 65.3 | Hurt |
| reduce slice size from full frame (104ms) to 1/5 frame (21ms) | SPVNAS · SemanticKITTI val | mIoU: 64.7 -> 57.8 | Hurt |
| reduce slice size from full frame (104ms) to 1/5 frame (21ms) | Helix4D · SemanticKITTI val | mIoU: 66.7 -> 66.8 | No clear effect |
| reduce slice size from full frame (104ms) to 1/5 frame (21ms) | Cylinder3D · SemanticKITTI val | mIoU: 66.9 -> 65.3 | Hurt |
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.
- PolarNet — PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020) · better on 4 of 4 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
- PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020 · similarity 0.87)
- Less Is More: Reducing Task and Model Complexity for 3D Point Cloud Semantic Segmentation (CVPR 2023 · similarity 0.87)
- Zero-Shot 4D Lidar Panoptic Segmentation (CVPR 2025 · similarity 0.87)
- MCD: Diverse Large-Scale Multi-Campus Dataset for Robot Perception (CVPR 2024 · similarity 0.85)
- D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation (CVPR 2025 · similarity 0.85)
- Multi-Space Alignments Towards Universal LiDAR Segmentation (CVPR 2024 · similarity 0.85)
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.
