Research map / Point clouds / Shape and surface completion
LiNeXt: Revisiting LiDAR Completion with Efficient Non-Diffusion Architectures
Wenzhe He, Xiaojun Chen, Ruiqi Wang, Ruihui Li, Huilong Pi, Jiapeng Zhang, Zhuo Tang, Kenli Li
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 |
|---|---|---|---|
| use LiNeXt non-diffusion network instead of LiDiff | scene completion · SemanticKITTI val | Chamfer Distance: 0.434 -> 0.214 | Improved |
| use LiNeXt instead of LiDiff | scene completion · SemanticKITTI efficiency | inference time (s): 33.359 -> 0.167 | Improved |
| add Refine Module (ablation) | LiNeXt · SemanticKITTI val | Chamfer Distance: 0.214 -> 0.149 | Improved |
| add Refine Module (ablation) | LiNeXt · SemanticKITTI efficiency | inference time (s): 0.167 -> 0.434 | Hurt |
| add Distance-aware Selected Repeat (ablation) | LiNeXt · SemanticKITTI val | Chamfer Distance: 0.215 -> 0.214 | No clear effect |
| add Distance-aware Selected Repeat (ablation) | LiNeXt · SemanticKITTI val | Voxel IoU @0.1m (%): 6.60 -> 6.30 | Hurt |
| add Multi-Scale Sparse Convolution (ablation) | LiNeXt · SemanticKITTI val | Voxel IoU @0.5m (%): 39.87 -> 41.07 | Improved |
| add Cross-Point Attention in place of standard cross-attention (ablation) | LiNeXt · SemanticKITTI val | Chamfer Distance: 0.227 -> 0.214 | Improved |
| evaluate on KITTI-360 without fine-tuning | LiNeXt with refinement · KITTI-360 seq 00 | Chamfer Distance: 0.149 -> 0.149 | No clear effect |
Closest papers by meaning
- Scaling Diffusion Models to Real-World 3D LiDAR Scene Completion (CVPR 2024 · similarity 0.92)
- Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion (ICCV 2025 · similarity 0.90)
- PatchScene: Patch-based Voxel Diffusion Model for Large-Scale Scene Completion (CVPR 2026 · similarity 0.90)
- Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion (AAAI 2021 · similarity 0.89)
- Flexible Depth Completion for Sparse and Varying Point Densities (CVPR 2024 · similarity 0.88)
- LiDAR-Net: A Real-scanned 3D Point Cloud Dataset for Indoor Scenes (CVPR 2024 · similarity 0.87)
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.
