.png){kind=icon}
.png){kind=icon}
Return
ExpPoint-MAE: Better Interpretability and Performance for Self-Supervised Point Cloud Transformers
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3388155.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper we delve into the properties of transformers, attained through self-supervision, in the point cloud domain. Specifically, we evaluate the effectiveness of Masked Autoencoding as a pretraining scheme, and explore Momentum Contrast as an alternative. In our study we investigate the impact of data quantity on the learned features, and uncover similarities in the transformer's behavior across domains. Through comprehensive visualizations, we observe that the transformer learns to attend to semantically meaningful regions, indicating that pretraining leads to a better understanding of the underlying geometry. Moreover, we examine the finetuning process and its effect on the learned representations. Based on that, we devise an unfreezing strategy which consistently outperforms our baseline without introducing any other modifications to the model or the training pipeline, and achieve state-of-the-art results in the classification task among transformer models.
Keywords:
Transformers
Solid modeling
Task analysis
Point cloud compression
Feature extraction
Training
Deep learning
Self-supervised learning
explainability
point clouds
self-supervision
transformers
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W