Face Alignment Across Large Poses: A 3D Solution

TL;DR

This paper introduces 3D Dense Face Alignment (3DDFA), a CNN-based framework that effectively aligns faces in large poses up to 90 degrees, overcoming challenges of visibility, appearance variation, and data labeling.

Contribution

The paper presents a novel 3D face model fitting framework and a data synthesis method for large-pose face alignment, addressing key limitations of existing approaches.

Findings

Significant improvement over state-of-the-art methods on AFLW database

Effective handling of large pose variations up to 90 degrees

Robust face alignment in profile and extreme poses

Abstract

Face alignment, which fits a face model to an image and extracts the semantic meanings of facial pixels, has been an important topic in CV community. However, most algorithms are designed for faces in small to medium poses (below 45 degree), lacking the ability to align faces in large poses up to 90 degree. The challenges are three-fold: Firstly, the commonly used landmark-based face model assumes that all the landmarks are visible and is therefore not suitable for profile views. Secondly, the face appearance varies more dramatically across large poses, ranging from frontal view to profile view. Thirdly, labelling landmarks in large poses is extremely challenging since the invisible landmarks have to be guessed. In this paper, we propose a solution to the three problems in an new alignment framework, called 3D Dense Face Alignment (3DDFA), in which a dense 3D face model is fitted to the image via convolutional neutral network (CNN). We also propose a method to synthesize large-scale training samples in profile views to solve the third problem of data labelling. Experiments on the challenging AFLW database show that our approach achieves significant improvements over state-of-the-art methods.