MANGO: Learning Disentangled Image Transformation Manifolds with Grouped Operators

TL;DR

MANGO introduces a method to learn disentangled, semantically meaningful image transformations using grouped operators, enabling efficient, composable transformations with a significantly faster training routine.

Contribution

The paper presents MANGO, a novel approach that learns disentangled image transformation operators in separate latent spaces, allowing user-defined transformations and a 100x faster training process.

Findings

Enables composition of image transformations.

Achieves 100x speedup over prior methods.

Learns semantically meaningful, disentangled operators.

Abstract

Learning semantically meaningful image transformations (i.e. rotation, thickness, blur) directly from examples can be a challenging task. Recently, the Manifold Autoencoder (MAE) proposed using a set of Lie group operators to learn image transformations directly from examples. However, this approach has limitations, as the learned operators are not guaranteed to be disentangled and the training routine is prohibitively expensive when scaling up the model. To address these limitations, we propose MANGO (transformation Manifolds with Grouped Operators) for learning disentangled operators that describe image transformations in distinct latent subspaces. Moreover, our approach allows practitioners the ability to define which transformations they aim to model, thus improving the semantic meaning of the learned operators. Through our experiments, we demonstrate that MANGO enables composition of image transformations and introduces a one-phase training routine that leads to a 100x speedup over prior works.