3D Dense Face Alignment via Graph Convolution Networks

TL;DR

This paper introduces a graph convolution network for 3D dense face alignment that directly learns features on 3D face meshes, effectively preserving geometric details and outperforming existing methods.

Contribution

The paper presents a novel graph convolution network that directly regresses 3D face coordinates on mesh structures, enhancing geometric detail preservation.

Findings

Outperforms state-of-the-art methods on multiple datasets

Effectively preserves geometric structure and details

Demonstrates superior accuracy in 3D face reconstruction

Abstract

Recently, 3D face reconstruction and face alignment tasks are gradually combined into one task: 3D dense face alignment. Its goal is to reconstruct the 3D geometric structure of face with pose information. In this paper, we propose a graph convolution network to regress 3D face coordinates. Our method directly performs feature learning on the 3D face mesh, where the geometric structure and details are well preserved. Extensive experiments show that our approach gains superior performance over state-of-the-art methods on several challenging datasets.