A Recurrent Encoder-Decoder Network for Sequential Face Alignment

TL;DR

This paper introduces a recurrent encoder-decoder network for real-time video face alignment, leveraging spatial and temporal recurrent learning to improve accuracy and generalization in facial point detection.

Contribution

It presents a novel model that combines spatial feedback loops and temporal feature decoupling with recurrent learning for enhanced face alignment.

Findings

Outperforms state-of-the-art methods on standard datasets.

Enables iterative coarse-to-fine alignment with a single network.

Improves generalization by decoupling pose/expression from identity.

Abstract

We propose a novel recurrent encoder-decoder network model for real-time video-based face alignment. Our proposed model predicts 2D facial point maps regularized by a regression loss, while uniquely exploiting recurrent learning at both spatial and temporal dimensions. At the spatial level, we add a feedback loop connection between the combined output response map and the input, in order to enable iterative coarse-to-fine face alignment using a single network model. At the temporal level, we first decouple the features in the bottleneck of the network into temporal-variant factors, such as pose and expression, and temporal-invariant factors, such as identity information. Temporal recurrent learning is then applied to the decoupled temporal-variant features, yielding better generalization and significantly more accurate results at test time. We perform a comprehensive experimental analysis, showing the importance of each component of our proposed model, as well as superior results over the state-of-the-art in standard datasets.