Unsupervised 3D Learning for Shape Analysis via Multiresolution Instance Discrimination

TL;DR

This paper introduces an unsupervised 3D shape encoding network that effectively performs various shape analysis tasks, achieving results comparable or superior to supervised methods, especially with limited labeled data.

Contribution

It proposes a novel multiresolution instance discrimination loss and adapts HR-Net for joint shape and point feature encoding from unlabeled 3D point clouds.

Findings

Outperforms all existing unsupervised methods in shape analysis tasks.

Achieves competitive results to supervised methods with minimal labeled data.

Surpasses supervised methods in fine-grained shape segmentation.

Abstract

Although unsupervised feature learning has demonstrated its advantages to reducing the workload of data labeling and network design in many fields, existing unsupervised 3D learning methods still cannot offer a generic network for various shape analysis tasks with competitive performance to supervised methods. In this paper, we propose an unsupervised method for learning a generic and efficient shape encoding network for different shape analysis tasks. The key idea of our method is to jointly encode and learn shape and point features from unlabeled 3D point clouds. For this purpose, we adapt HR-Net to octree-based convolutional neural networks for jointly encoding shape and point features with fused multiresolution subnetworks and design a simple-yet-efficient Multiresolution Instance Discrimination (MID) loss for jointly learning the shape and point features. Our network takes a 3D point cloud as input and output both shape and point features. After training, the network is concatenated with simple task-specific back-end layers and fine-tuned for different shape analysis tasks. We evaluate the efficacy and generality of our method and validate our network and loss design with a set of shape analysis tasks, including shape classification, semantic shape segmentation, as well as shape registration tasks. With simple back-ends, our network demonstrates the best performance among all unsupervised methods and achieves competitive performance to supervised methods, especially in tasks with a small labeled dataset. For fine-grained shape segmentation, our method even surpasses existing supervised methods by a large margin.