Joint Data and Feature Augmentation for Self-Supervised Representation Learning on Point Clouds

TL;DR

This paper introduces a novel self-supervised learning framework for point clouds that fuses data and feature augmentations, improving transferability and achieving state-of-the-art results in classification and segmentation tasks.

Contribution

It proposes a combined augmentation framework with new data and feature augmentation methods, enhancing representation transferability in self-supervised point cloud learning.

Findings

Achieves state-of-the-art results in object classification

Demonstrates effective transferability to segmentation tasks

Validates the framework's robustness across datasets

Abstract

To deal with the exhausting annotations, self-supervised representation learning from unlabeled point clouds has drawn much attention, especially centered on augmentation-based contrastive methods. However, specific augmentations hardly produce sufficient transferability to high-level tasks on different datasets. Besides, augmentations on point clouds may also change underlying semantics. To address the issues, we propose a simple but efficient augmentation fusion contrastive learning framework to combine data augmentations in Euclidean space and feature augmentations in feature space. In particular, we propose a data augmentation method based on sampling and graph generation. Meanwhile, we design a data augmentation network to enable a correspondence of representations by maximizing consistency between augmented graph pairs. We further design a feature augmentation network that encourages the model to learn representations invariant to the perturbations using an encoder perturbation. We comprehensively conduct extensive object classification experiments and object part segmentation experiments to validate the transferability of the proposed framework. Experimental results demonstrate that the proposed framework is effective to learn the point cloud representation in a self-supervised manner, and yields state-of-the-art results in the community. The source code is publicly available at: https://zhiyongsu.github.io/Project/AFSRL.html.