AI Paper Atlas

Research map / Point clouds / Point cloud classification

Masked Autoencoders for Point Cloud Self-Supervised Learning

Yatian Pang, Wenxiao Wang, Francis E.H. Tay, Wei Liu, Yonghong Tian, Li Yuan

ECCV 2022 · Point clouds › Point cloud classification · approach: Masked point autoencoders · 695 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 Point-MAE pretraining vs training from scratchTransformer · ScanObjectNN PB-T50-RSaccuracy: 77.24 -> 85.18Improved
add Point-MAE pretraining vs training from scratchTransformer · ScanObjectNN OBJ-BGaccuracy: 79.86 -> 90.02Improved
add Point-MAE pretraining vs training from scratchTransformer · ModelNet40accuracy: 91.4% -> 93.8%Improved
increase input points from 1024 to 8192Point-MAE · ModelNet40accuracy: 93.8% -> 94.04%Improved
replace Point-BERT pretraining with Point-MAETransformer · ScanObjectNN PB-T50-RSaccuracy: 83.07 -> 85.18Improved
replace Point-BERT pretraining with Point-MAETransformer · ModelNet40accuracy: 93.2% -> 93.8%Improved
replace Point-BERT with Point-MAE for few-shotTransformer · ModelNet40 few-shotaccuracy: 92.7 -> 95.0Improved
add Point-MAE pretraining on part segmentationTransformer · ShapeNetPartmIoU_I: 85.1 -> 86.1Improved
use random masking instead of block maskingPoint-MAE · ModelNet40accuracy: 92.67 -> 93.19Improved
raise random masking ratio from 40% to 60% (ablation)Point-MAE · ModelNet40accuracy: 92.46 -> 93.19Improved
raise random masking ratio from 60% to 90% (ablation)Point-MAE · ModelNet40accuracy: 93.19 -> 92.63Hurt
shift mask tokens to the lightweight decoder (ablation)Point-MAE · ModelNet40accuracy: 92.14% -> 93.19%Improved

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