Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network

TL;DR

This paper introduces a fast, lightweight neural network that jointly reconstructs 3D faces and performs dense alignment from a single image using a novel UV position map representation, surpassing state-of-the-art methods.

Contribution

It presents a simple, model-free approach with a novel UV position map and a weight mask, enabling accurate 3D face reconstruction and alignment with high speed.

Findings

Outperforms existing methods on multiple datasets

Reconstructs full facial geometry with semantic meaning

Processes images in only 9.8 milliseconds

Abstract

We propose a straightforward method that simultaneously reconstructs the 3D facial structure and provides dense alignment. To achieve this, we design a 2D representation called UV position map which records the 3D shape of a complete face in UV space, then train a simple Convolutional Neural Network to regress it from a single 2D image. We also integrate a weight mask into the loss function during training to improve the performance of the network. Our method does not rely on any prior face model, and can reconstruct full facial geometry along with semantic meaning. Meanwhile, our network is very light-weighted and spends only 9.8ms to process an image, which is extremely faster than previous works. Experiments on multiple challenging datasets show that our method surpasses other state-of-the-art methods on both reconstruction and alignment tasks by a large margin.