Face Alignment using a 3D Deeply-initialized Ensemble of Regression Trees
It addresses robust face alignment for computer vision applications, with incremental improvements over existing methods.
The paper tackles face alignment under occlusions, deformations, and pose variations by proposing 3DDE, a method that initializes with a 3D face model and uses a cascade of regression trees, improving state-of-the-art on multiple datasets like 300W and COFW.
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.