Mitigating Positional Leakage in 3D Masked Autoencoders for Robust Representation Learning

2026-06-30

AI Summary

This work addresses the limitation of existing 3D Masked Autoencoders (MAEs), which overly rely on positional information during spatial coordinate reconstruction, thereby compromising semantic representation learning. To mitigate this “positional leakage” issue, the authors propose the MPL-MAE framework, featuring two key innovations: a novel positional embedding module that suppresses metric-dominant signals while preserving geometric topology, and a gated positional interface that dynamically modulates the injection of positional cues during reconstruction. By effectively balancing the interplay between spatial priors and semantic features, the proposed method significantly enhances the robustness and informativeness of learned representations across diverse downstream tasks, demonstrating its superiority over prior approaches.

Abstract

Masked autoencoding has emerged as a prominent paradigm for self-supervised learning on 3D point clouds, achieving competitive performance across downstream tasks. Unlike its 2D counterpart, 3D masked autoencoding directly reconstructs spatial coordinates, making it inherently susceptible to positional leakage. In this work, we identify that the decoder in existing 3D MAE frameworks tends to over-rely on positional information, which weakens semantic representation learning and leads to suboptimal feature quality. To address this issue, we propose MPL-MAE, a masked point learning framework that mitigates positional over-reliance while enhancing the utilization of encoder features. Specifically, we introduce a recalibrated positional embedding module that suppresses metric-dominant coordinate signals while preserving geometric topology, together with a gated positional interface module that dynamically regulates positional injection during reconstruction. These designs promote a more balanced interaction between spatial priors and semantic features, yielding robust and informative representations. Extensive experiments across downstream tasks demonstrate that MPL-MAE consistently achieves competitive performance, validating its effectiveness. Code is available at https://github.com/yanx57/MPL-MAE.

Problem

• positional leakage
• 3D masked autoencoders
• representation learning
• semantic features
• point cloud

Innovation

• positional leakage
• masked autoencoder
• 3D point clouds
• self-supervised learning
• representation learning