Research map / Point clouds / Point cloud classification
CN: Channel Normalization For Point Cloud Recognition
Zetong Yang, Yanan Sun, Shu Liu, Xiaojuan Qi, Jiaya Jia
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 channel normalization (SSG) | PointNet++ · ModelNet40 | accuracy (%): 90.7 -> 92.9 | Improved |
| add channel normalization (MSG) | PointNet++ · ModelNet40 | accuracy (%): 90.7 -> 93.3 | Improved |
| use CN instead of RS-CNN fusion (SSG) | PointNet++ SSG · ModelNet40 | accuracy (%): 92.2 -> 92.9 | Improved |
| add channel normalization to SA layers | PointNet++ · ShapeNet part segmentation | class mIoU (%): 81.9 -> 83.9 | Improved |
| add two long-range skip connections to CN (CN2) | CN · ShapeNet part segmentation | class mIoU (%): 83.9 -> 84.3 | Improved |
| add channel normalization | 3DSSD · KITTI test Car AP_3D Easy | AP (%): 88.36 -> 90.55 | Improved |
| add channel normalization | 3DSSD · KITTI test Car AP_BEV Easy | AP (%): 92.66 -> 94.51 | Improved |
| add channel normalization | 3DSSD · KITTI test Car AP_BEV Hard | AP (%): 85.86 -> 85.86 | No clear effect |
| add channel normalization | 3DSSD · KITTI test Car AP_3D Moderate | AP (%): 79.57 -> 79.89 | Improved |
| use CN instead of RS-CNN fusion layers | 3DSSD · KITTI val Car Moderate | 3D mAP (%): 82.88 -> 85.01 | Improved |
| use CN instead of PointNet++ SA layers | 3DSSD · KITTI val Car Moderate | 3D mAP (%): 82.37 -> 85.01 | Improved |
| add learnable condition parameters alpha1 and alpha2 to CN (ablation) | 3DSSD + CN · KITTI val Car Moderate | 3D mAP (%): 84.60 -> 85.01 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- 3DSSD — 3DSSD: Point-Based 3D Single Stage Object Detector (CVPR 2020) · better on 5 of 6 shared results
- PointConv — PointConv: Deep Convolutional Networks on 3D Point Clouds (CVPR 2019) · better on 1 of 1 shared result
- PointCNN — PointCNN: Convolution On X-Transformed Points (NeurIPS 2018) · better on 1 of 1 shared result
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
- GraphFit: Learning Multi-Scale Graph-Convolutional Representation for Point Cloud Normal Estimation (ECCV 2022 · similarity 0.87)
- Geometry and Learning Co-Supported Normal Estimation for Unstructured Point Cloud (CVPR 2020 · similarity 0.86)
- Improving 3D Object Detection With Channel-Wise Transformer (ICCV 2021 · similarity 0.85)
- X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition (CVPR 2024 · similarity 0.85)
- An Explicit Frame Construction for Normalizing 3D Point Clouds (ICML 2024 · similarity 0.84)
- Occlusion-aware Text-Image-Point Cloud Pretraining for Open-World 3D Object Recognition (CVPR 2025 · similarity 0.84)
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.
