AI Paper Atlas

Research map / Point clouds / Point cloud classification

Contrastive Predictive Autoencoders for Dynamic Point Cloud Self-Supervised Learning

Xiaoxiao Sheng, Zhiqiang Shen, Gang Xiao

AAAI 2023 · Point clouds › Point cloud classification · approach: Masked point autoencoders · 16 citations (OpenAlex)

Read the paper

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
Pretrain with a longer sequence (24 rather than 12 frames) (ablation)CPR · MSRAction3D, finetune 24 framesaccuracy (%): 91.99 -> 93.38Improved
Reconstruct a segment rather than one frame (ablation)CPR reconstruction · MSRAction3Daccuracy (%): 91.32 -> 92.68Improved
Colorize the reconstructed segment (ablation)CPR reconstruction · MSRAction3Daccuracy (%): 92.68 -> 93.38Improved
Increase reconstruction density from 256 to 512 points per frame (ablation)CPR reconstruction · MSRAction3Daccuracy (%): 93.38 -> 91.99Hurt
Increase reconstruction density from 256 to 1024 points per frame (ablation)CPR reconstruction · MSRAction3Daccuracy (%): 93.38 -> 92.36Hurt
Add temporally adjacent hard negatives (ablation)Local contrastive prediction · MSRAction3Daccuracy (%): 91.34 -> 92.23Improved
Add global contrastive prediction to local contrastive prediction (ablation)CPR · MSRAction3Daccuracy (%): 91.34 -> 92.55Improved
Add sequence reconstruction to contrastive prediction (ablation)CPR · MSRAction3Daccuracy (%): 92.55 -> 93.38Improved
Use contrastive prediction and reconstruction together rather than sequence reconstruction alone (ablation)CPR · MSRAction3Daccuracy (%): 87.10 -> 93.38Improved
Transfer CPR pretraining from NTU-RGBD 60 rather than MSRAction3DCPR · MSRAction3D, semi-supervision with 30% MSRaccuracy (%): 86.4 -> 87.2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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