Multi-modal Multi-task Pre-training for Improved Point Cloud Understanding

TL;DR

This paper introduces MMPT, a multi-task pre-training framework that leverages three self-supervised tasks across 3D point clouds and 2D images to improve understanding without needing 3D annotations.

Contribution

The paper presents a novel multi-task pre-training approach combining token-level, point-level, and contrastive learning for enhanced 3D point cloud understanding.

Findings

Outperforms state-of-the-art methods on multiple benchmarks.

Effective in both discriminant and generative tasks.

Operates without requiring 3D annotations.

Abstract

Recent advances in multi-modal pre-training methods have shown promising effectiveness in learning 3D representations by aligning multi-modal features between 3D shapes and their corresponding 2D counterparts. However, existing multi-modal pre-training frameworks primarily rely on a single pre-training task to gather multi-modal data in 3D applications. This limitation prevents the models from obtaining the abundant information provided by other relevant tasks, which can hinder their performance in downstream tasks, particularly in complex and diverse domains. In order to tackle this issue, we propose MMPT, a Multi-modal Multi-task Pre-training framework designed to enhance point cloud understanding. Specifically, three pre-training tasks are devised: (i) Token-level reconstruction (TLR) aims to recover masked point tokens, endowing the model with representative learning abilities. (ii) Point-level reconstruction (PLR) is integrated to predict the masked point positions directly, and the reconstructed point cloud can be considered as a transformed point cloud used in the subsequent task. (iii) Multi-modal contrastive learning (MCL) combines feature correspondences within and across modalities, thus assembling a rich learning signal from both 3D point cloud and 2D image modalities in a self-supervised manner. Moreover, this framework operates without requiring any 3D annotations, making it scalable for use with large datasets. The trained encoder can be effectively transferred to various downstream tasks. To demonstrate its effectiveness, we evaluated its performance compared to state-of-the-art methods in various discriminant and generative applications under widely-used benchmarks.