Data Efficient 3D Learner via Knowledge Transferred from 2D Model

TL;DR

This paper introduces a method to improve 3D learning efficiency by transferring knowledge from 2D models using RGB-D images, enabling better performance with limited 3D data.

Contribution

The authors propose a novel approach that leverages 2D semantic segmentation models to augment RGB-D data for pre-training 3D models, enhancing data efficiency and semi-supervised learning.

Findings

Outperforms existing methods on 3D label efficiency tasks

Achieves new state-of-the-art results on ScanNet

Improves semi-supervised learning performance

Abstract

Collecting and labeling the registered 3D point cloud is costly. As a result, 3D resources for training are typically limited in quantity compared to the 2D images counterpart. In this work, we deal with the data scarcity challenge of 3D tasks by transferring knowledge from strong 2D models via RGB-D images. Specifically, we utilize a strong and well-trained semantic segmentation model for 2D images to augment RGB-D images with pseudo-label. The augmented dataset can then be used to pre-train 3D models. Finally, by simply fine-tuning on a few labeled 3D instances, our method already outperforms existing state-of-the-art that is tailored for 3D label efficiency. We also show that the results of mean-teacher and entropy minimization can be improved by our pre-training, suggesting that the transferred knowledge is helpful in semi-supervised setting. We verify the effectiveness of our approach on two popular 3D models and three different tasks. On ScanNet official evaluation, we establish new state-of-the-art semantic segmentation results on the data-efficient track.