Image Shape Manipulation from a Single Augmented Training Sample

TL;DR

DeepSIM is a generative model enabling complex image shape manipulations from a single sample by leveraging extensive augmentation and primitive representations, achieving high-quality results.

Contribution

Introducing DeepSIM, a novel single-image training framework that uses augmentation and primitive representations for flexible image shape manipulation.

Findings

Effective single-image training with extensive augmentation.

Primitive representations influence manipulation ease and expressiveness.

High-quality image manipulation results demonstrated.

Abstract

In this paper, we present DeepSIM, a generative model for conditional image manipulation based on a single image. We find that extensive augmentation is key for enabling single image training, and incorporate the use of thin-plate-spline (TPS) as an effective augmentation. Our network learns to map between a primitive representation of the image to the image itself. The choice of a primitive representation has an impact on the ease and expressiveness of the manipulations and can be automatic (e.g. edges), manual (e.g. segmentation) or hybrid such as edges on top of segmentations. At manipulation time, our generator allows for making complex image changes by modifying the primitive input representation and mapping it through the network. Our method is shown to achieve remarkable performance on image manipulation tasks.