Inverting Generative Adversarial Renderer for Face Reconstruction

TL;DR

This paper introduces a Generative Adversarial Renderer (GAR) that models realistic face images, improving 3D face reconstruction accuracy by reducing domain-shift noise and enabling more effective optimization.

Contribution

The work presents a novel neural renderer that produces realistic images, and a new face reconstruction pipeline that inverts this renderer for improved 3D face parameter estimation.

Findings

Achieves state-of-the-art results on multiple datasets.

Produces more realistic face images compared to traditional differentiable renderers.

Reduces domain-shift noise in face reconstruction processes.

Abstract

Given a monocular face image as input, 3D face geometry reconstruction aims to recover a corresponding 3D face mesh. Recently, both optimization-based and learning-based face reconstruction methods have taken advantage of the emerging differentiable renderer and shown promising results. However, the differentiable renderer, mainly based on graphics rules, simplifies the realistic mechanism of the illumination, reflection, \etc, of the real world, thus cannot produce realistic images. This brings a lot of domain-shift noise to the optimization or training process. In this work, we introduce a novel Generative Adversarial Renderer (GAR) and propose to tailor its inverted version to the general fitting pipeline, to tackle the above problem. Specifically, the carefully designed neural renderer takes a face normal map and a latent code representing other factors as inputs and renders a realistic face image. Since the GAR learns to model the complicated real-world image, instead of relying on the simplified graphics rules, it is capable of producing realistic images, which essentially inhibits the domain-shift noise in training and optimization. Equipped with the elaborated GAR, we further proposed a novel approach to predict 3D face parameters, in which we first obtain fine initial parameters via Renderer Inverting and then refine it with gradient-based optimizers. Extensive experiments have been conducted to demonstrate the effectiveness of the proposed generative adversarial renderer and the novel optimization-based face reconstruction framework. Our method achieves state-of-the-art performances on multiple face reconstruction datasets.