AI Paper Atlas

Research map / Point clouds / Segmentation and detection

PolarMix: A General Data Augmentation Technique for LiDAR Point Clouds

Aoran Xiao, Jiaxing Huang, Dayan Guan, Kaiwen Cui, Shijian Lu, Ling Shao

NeurIPS 2022 · Point clouds › Segmentation and detection · approach: LiDAR segmentation for driving · 37 citations (OpenAlex)

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 changedWhereMeasuredEffect
Add PolarMix to MinkNet trainingMinkNet · SemanticKITTI valmIoU: 55.9 -> 65.0Improved
Add PolarMix to SPVCNN trainingSPVCNN · SemanticKITTI valmIoU: 58.0 -> 66.2Improved
Add PolarMix to MinkNet trainingMinkNet · nuScenes-lidarseg valmIoU: 67.1 -> 72.0Improved
Add PolarMix to MinkNet trainingMinkNet · SemanticPOSS valmIoU: 52.1 -> 57.4Improved
Add PolarMix to SPVCNN trainingSPVCNN · nuScenes-lidarseg valmIoU: 68.4 -> 72.1Improved
Add PolarMix to SPVCNN trainingSPVCNN · SemanticPOSS valmIoU: 50.7 -> 58.6Improved
Add PolarMixRandLA-Net · SemanticKITTI val (10% training data)mIoU: 45.4 -> 50.5Improved
Add PolarMixCylinder3D · SemanticKITTI val (10% training data)mIoU: 60.6 -> 62.5Improved
Add PolarMix augmentationPointPillar, Second, and CenterNet · nuScenes valmAP: 41.8 -> 43.7; 48.8 -> 50.3; 52.8 -> 55.4Improved
Mix labeled source scans with pseudo-labeled target scans using PolarMixMinkNet self-training · SynLiDAR → SemanticKITTImIoU: 28.9 -> 31.0Improved
Mix labeled source scans with pseudo-labeled target scans using PolarMixMinkNet self-training · SynLiDAR → SemanticPOSSmIoU: 29.6 -> 30.4Improved
Add scene-level swapping (ablation)SPVCNN · SemanticKITTI val (sequence 00 training)mIoU: 48.9 -> 50.8Improved

Later papers that compared with this method

1 later accepted paper reports PolarMix as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy