Deep Active Latent Surfaces for Medical Geometries

TL;DR

This paper introduces a hybrid deep learning approach for 3D shape reconstruction in medical imaging, using vertex-wise latent vectors with regularization to balance flexibility and overfitting.

Contribution

It proposes a novel shape representation with vertex-specific latent vectors constrained during training and independently updated during inference, enhancing generalization.

Findings

Effective in reconstructing medical geometries from noisy data

Balances flexibility and overfitting through regularized latent vectors

Demonstrates improved performance on medical image tasks

Abstract

Shape priors have long been known to be effective when reconstructing 3D shapes from noisy or incomplete data. When using a deep-learning based shape representation, this often involves learning a latent representation, which can be either in the form of a single global vector or of multiple local ones. The latter allows more flexibility but is prone to overfitting. In this paper, we advocate a hybrid approach representing shapes in terms of 3D meshes with a separate latent vector at each vertex. During training the latent vectors are constrained to have the same value, which avoids overfitting. For inference, the latent vectors are updated independently while imposing spatial regularization constraints. We show that this gives us both flexibility and generalization capabilities, which we demonstrate on several medical image processing tasks.