Face Alignment using a 3D Deeply-initialized Ensemble of Regression Trees

Abstract

Face alignment algorithms locate a set of landmark points in images of faces taken in unrestricted situations. State-of-the-art approaches typically fail or lose accuracy in the presence of occlusions, strong deformations, large pose variations and ambiguous configurations. In this paper we present 3DDE, a robust and efficient face alignment algorithm based on a coarse-to-fine cascade of ensembles of regression trees. It is initialized by robustly fitting a 3D face model to the probability maps produced by a convolutional neural network. With this initialization we address self-occlusions and large face rotations. Further, the regressor implicitly imposes a prior face shape on the solution, addressing occlusions and ambiguous face configurations. Its coarse-to-fine structure tackles the combinatorial explosion of parts deformation. In the experiments performed, 3DDE improves the state-of-the-art in 300W, COFW, AFLW and WFLW data sets. Finally, we perform cross-dataset experiments that reveal the existence of a significant data set bias in these benchmarks.

arXiv commentAccepted Version to Computer Vision and Image Understanding

References in corpus (5)

• Face Alignment Across Large Poses: A 3D Solution
• Face Alignment in Full Pose Range: A 3D Total Solution
• Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources
• Face Alignment Using K-Cluster Regression Forests With Weighted Splitting
• Deep Deformation Network for Object Landmark Localization

Cited by in corpus (6)

• Pixel-in-Pixel Net: Towards Efficient Facial Landmark Detection in the Wild
• Multi-task head pose estimation in-the-wild
• Heatmap Regression via Randomized Rounding
• A Vector-based Representation to Enhance Head Pose Estimation
• Multistage Model for Robust Face Alignment Using Deep Neural Networks
• Human Recognition Using Face in Computed Tomography