AI Paper Atlas

Research map / Point clouds / Point cloud classification

GD-MAE: Generative Decoder for MAE Pre-Training on LiDAR Point Clouds

Honghui Yang, Tong He, Jiaheng Liu, Hua Chen, Boxi Wu, Binbin Lin, Xiaofei He, Wanli Ouyang

CVPR 2023 · Point clouds › Point cloud classification · approach: Masked point autoencoders · 72 citations (OpenAlex)

This paper tried a common change: Train longer

Increase pre-training epochs from 30 to 120 · Waymo validation · Vehicle AP: 66.54 -> 66.89 Improved

What 387 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

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
Use GD-MAE pre-training (ablation)SPT detector · ONCE validationmAP: 62.62 -> 64.92 (+2.30)Improved
Use GD-MAE pre-training (ablation)Waymo detector · Waymo validation, 5% labeled dataVehicle mAPH/L2: 59.97 -> 63.86; +3.89Improved
Add a pyramid architecture (ablation)SPT detector · Waymo validationVehicle APH/L2: 62.10/61.61 -> 64.20/63.70Improved
Add a shortcut connection (ablation)Pyramid SPT · Waymo validationVehicle APH/L2: 64.20/63.70 -> 66.02/65.55Improved
Add GD-MAE pre-training (ablation)Pyramid SPT with shortcut · Waymo validationVehicle APH/L2: 66.02/65.55 -> 67.00/66.54Improved
Use patch-wise rather than block-wise masking (ablation)GD-MAE · Waymo validation, 5% labeled dataVehicle AP: 61.60 -> 62.65Improved
Use point-wise rather than patch-wise masking (ablation)GD-MAE · Waymo validation, 5% labeled dataVehicle AP: 62.65 -> 62.41Hurt
Replace the transformer decoder with the generative decoder (ablation)MAE pre-training · Waymo validationVehicle APH/L2: 66.24 -> 66.54Improved
Replace the transformer decoder with the generative decoder (ablation)MAE pre-training · Waymo decoder runtimeruntime: 27.1ms -> 3.2msImproved
Increase the pre-training data from 20% to 100% (ablation)GD-MAE · Waymo validationmAPH/L2: 67.14 -> 67.64Improved
Increase the labeled-data fraction from 20% to 100% (ablation)GD-MAE fine-tuning · Waymo validationVehicle mAPH/L2: 67.14 -> 67.64No clear effect
Increase pre-training epochs from 30 to 120 (ablation)GD-MAE · Waymo validationVehicle AP: 66.54 -> 66.89Improved

Later papers that compared with this method

2 later accepted papers report GD-MAE as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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