Learning Hyperbolic Representations for Unsupervised 3D Segmentation

Joy Hsu, Jeffrey Gu, Gong Her Wu, Wah Chiu, Serena Yeung

Keywords: unsupervised learning, representation learning, segmentation, biocomputing

Abstract: There exists a need for unsupervised 3D segmentation on complex volumetric data, particularly when annotation ability is limited or discovery of new categories is desired. Using the observation that much of 3D volumetric data is innately hierarchical, we propose learning effective representations of 3D patches for unsupervised segmentation through a variational autoencoder (VAE) with a hyperbolic latent space and a proposed gyroplane convolutional layer, which better models the underlying hierarchical structure within a 3D image. We also introduce a hierarchical triplet loss and multi-scale patch sampling scheme to embed relationships across varying levels of granularity. We demonstrate the effectiveness of our hyperbolic representations for unsupervised 3D segmentation on a hierarchical toy dataset, BraTS whole tumor dataset, and cryogenic electron microscopy data.

One-sentence Summary: We introduce an approach for learning hyperbolic embeddings of 3D data that can model inherent hierarchy within the data, and can be used to achieve state-of-the-art performance for unsupervised 3D segmentation.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=AeYDgs1MX7