AI Paper Atlas

Research map / Point clouds / Point cloud classification

Shape Self-Correction for Unsupervised Point Cloud Understanding

Ye Chen, Jinxian Liu, Bingbing Ni, Hang Wang, Jiancheng Yang, Ning Liu, Teng Li, Qi Tian

ICCV 2021 · Point clouds › Point cloud classification · approach: Self-supervised point cloud pretraining · 55 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 template (original shape) features (ablation)Shape Self-Correction · ModelNet40accuracy: 87.8 -> 92.4Improved
distort with exchange, replace, rotate, translate, scale instead of noise only (ablation)Shape Self-Correction · ModelNet40accuracy: 87.2 -> 92.4Improved
pre-train with Shape Self-Correction before fine-tuningPointNet · ShapeNetPart 1% dataIns. mIoU: 68.6 -> 72.9Improved
pre-train with Shape Self-Correction before fine-tuningRSCNN · ShapeNetPart 1% dataIns. mIoU: 71.6 -> 74.3Improved
pre-train with Shape Self-Correction before fine-tuningRSCNN · ShapeNetPart 100% dataIns. mIoU: 84.3 -> 85.2Improved
pre-train with Shape Self-Correction before fine-tuningRSCNN · ModelNet40accuracy: 91.7 -> 93.0Improved
pre-train with Shape Self-Correction before fine-tuningRSCNN · ScanNetaccuracy: 90.1 -> 92.9Improved
pre-train with Shape Self-Correction before fine-tuningPointNet · ScanNetaccuracy: 87.8 -> 89.7Improved
pre-train with PointGLR instead of random initRSCNN · ModelNet10accuracy: 94.8 -> 94.8No clear effect
use RSCNN instead of PointNet as encoderShape Self-Correction · ModelNet40 (unsupervised transfer)accuracy: 89.9 -> 92.4Improved
increase number of incorrect points K beyond 256 (ablation)Shape Self-Correction · ModelNet40accuracyNo clear effect

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