# Synthesizing Normalized Faces from Facial Identity Features

## TL;DR

This paper introduces a method to generate frontal, neutral-expression face images from input photos by leveraging invariant facial features, enabling applications like attribute analysis and 3D avatar creation.

## Contribution

The approach uses a novel invariant feature encoding and a decoder trained solely on frontal, neutral faces to synthesize normalized face images.

## Key findings

- Produces high-quality frontal face images from diverse inputs
- Enables applications like attribute analysis and avatar creation
- Uses invariant features for robust face normalization

## Abstract

We present a method for synthesizing a frontal, neutral-expression image of a person's face given an input face photograph. This is achieved by learning to generate facial landmarks and textures from features extracted from a facial-recognition network. Unlike previous approaches, our encoding feature vector is largely invariant to lighting, pose, and facial expression. Exploiting this invariance, we train our decoder network using only frontal, neutral-expression photographs. Since these photographs are well aligned, we can decompose them into a sparse set of landmark points and aligned texture maps. The decoder then predicts landmarks and textures independently and combines them using a differentiable image warping operation. The resulting images can be used for a number of applications, such as analyzing facial attributes, exposure and white balance adjustment, or creating a 3-D avatar.