Learning Hyperbolic Representations for Unsupervised 3D Segmentation

by   Joy Hsu, et al.

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.


page 5

page 6

page 8

page 14


APo-VAE: Text Generation in Hyperbolic Space

Natural language often exhibits inherent hierarchical structure ingraine...

Contrastive Multi-view Hyperbolic Hierarchical Clustering

Hierarchical clustering recursively partitions data at an increasingly f...

A Differentiable Gaussian-like Distribution on Hyperbolic Space for Gradient-Based Learning

Hyperbolic space is a geometry that is known to be well-suited for repre...

Unsupervised Hyperbolic Representation Learning via Message Passing Auto-Encoders

Most of the existing literature regarding hyperbolic embedding concentra...

Hierarchical Representations with Poincaré Variational Auto-Encoders

The Variational Auto-Encoder (VAE) model has become widely popular as a ...

Unsupervised Representations of Pollen in Bright-Field Microscopy

We present the first unsupervised deep learning method for pollen analys...

Corner-Sharing Tetrahedra for Modeling Micro-Structure

State-of-the-art representations of volumetric multi-scale shape and str...