Real-time Virtual-Try-On from a Single Example Image through Deep Inverse Graphics and Learned Differentiable Renderers

TL;DR

This paper introduces a real-time deep learning framework for virtual try-on that maps a single reference image into rendering parameters, using a trainable imitator for non-differentiable renderers, enabling practical AR applications.

Contribution

It presents a hybrid inverse graphics approach with a self-supervised learning method and a trainable imitator for non-differentiable renderers, facilitating real-time virtual try-on from a single image.

Findings

Achieves real-time performance on portable devices.

Accurately reproduces rendering behavior with the imitator network.

Enables virtual try-on for unknown products from a single reference image.

Abstract

Augmented reality applications have rapidly spread across online platforms, allowing consumers to virtually try-on a variety of products, such as makeup, hair dying, or shoes. However, parametrizing a renderer to synthesize realistic images of a given product remains a challenging task that requires expert knowledge. While recent work has introduced neural rendering methods for virtual try-on from example images, current approaches are based on large generative models that cannot be used in real-time on mobile devices. This calls for a hybrid method that combines the advantages of computer graphics and neural rendering approaches. In this paper we propose a novel framework based on deep learning to build a real-time inverse graphics encoder that learns to map a single example image into the parameter space of a given augmented reality rendering engine. Our method leverages self-supervised learning and does not require labeled training data which makes it extendable to many virtual try-on applications. Furthermore, most augmented reality renderers are not differentiable in practice due to algorithmic choices or implementation constraints to reach real-time on portable devices. To relax the need for a graphics-based differentiable renderer in inverse graphics problems, we introduce a trainable imitator module. Our imitator is a generative network that learns to accurately reproduce the behavior of a given non-differentiable renderer. We propose a novel rendering sensitivity loss to train the imitator, which ensures that the network learns an accurate and continuous representation for each rendering parameter. Our framework enables novel applications where consumers can virtually try-on a novel unknown product from an inspirational reference image on social media. It can also be used by graphics artists to automatically create realistic rendering from a reference product image.