Pose-Invariant Face Alignment with a Single CNN

TL;DR

This paper introduces a visualization layer for CNNs that improves pose-invariant face alignment, achieving state-of-the-art accuracy and faster training by enabling end-to-end optimization.

Contribution

The novel visualization layer allows joint optimization within CNNs, addressing previous limitations like slow training and handcrafted features in face alignment.

Findings

State-of-the-art accuracy on multiple datasets

Training time reduced by more than 50%

Effective across various CNN architectures

Abstract

Face alignment has witnessed substantial progress in the last decade. One of the recent focuses has been aligning a dense 3D face shape to face images with large head poses. The dominant technology used is based on the cascade of regressors, e.g., CNN, which has shown promising results. Nonetheless, the cascade of CNNs suffers from several drawbacks, e.g., lack of end-to-end training, hand-crafted features and slow training speed. To address these issues, we propose a new layer, named visualization layer, that can be integrated into the CNN architecture and enables joint optimization with different loss functions. Extensive evaluation of the proposed method on multiple datasets demonstrates state-of-the-art accuracy, while reducing the training time by more than half compared to the typical cascade of CNNs. In addition, we compare multiple CNN architectures with the visualization layer to further demonstrate the advantage of its utilization.