Quasi-Conformal Convolution : A Learnable Convolution for Deep Learning on Simply Connected Open Surfaces

TL;DR

This paper introduces Quasi-Conformal Convolution (QCC), a learnable convolution framework for deep learning on simply-connected open surfaces, enabling adaptive geometric analysis with applications in image classification and medical imaging.

Contribution

The paper proposes a novel QCC framework that leverages quasi-conformal mappings for adaptive, learnable convolution on non-Euclidean surfaces, unifying various spatial convolutions and enhancing geometric deep learning.

Findings

QCC improves classification accuracy on surface-based image data.

QCC enhances medical image analysis, such as craniofacial and lesion segmentation.

QCC demonstrates superior performance compared to existing methods.

Abstract

Deep learning on non-Euclidean domains is important for analyzing complex geometric data that lacks common coordinate systems and familiar Euclidean properties. A central challenge in this field is to define convolution on domains, which inherently possess irregular and non-Euclidean structures. In this work, we introduce Quasi-conformal Convolution (QCC), a novel framework for defining convolution on simply-connected open surfaces using quasi-conformal theories. Each QCC operator is linked to a specific quasi-conformal mapping, enabling the adjustment of the convolution operation through manipulation of this mapping. By utilizing trainable estimator modules that produce quasi-conformal mappings, QCC facilitates adaptive and learnable convolution operators that can be dynamically adjusted according to the underlying data structured on the surfaces. QCC unifies a broad range of spatially defined convolutions, facilitating the learning of tailored convolution operators on each underlying surface optimized for specific tasks. Building on this foundation, we develop the Quasi-Conformal Convolutional Neural Network (QCCNN) to address a variety of tasks related to geometric data. We validate the efficacy of QCCNN through the classification of images defined on curvilinear simply-connected open Riemann surfaces, demonstrating superior performance in this context. Additionally, we explore its potential in medical applications, including craniofacial analysis using 3D facial data and lesion segmentation on 3D human faces, achieving enhanced accuracy and reliability.