Recurrent Regression for Face Recognition

TL;DR

This paper introduces a recurrent regression neural network (RRNN) framework that unifies cross-pose and video-based face recognition by modeling sequential image dependencies and adaptively memorizing relevant information.

Contribution

The paper proposes a novel RRNN framework that explicitly models sequential dependencies for face recognition across poses and videos, improving recognition accuracy.

Findings

Effective in static face dataset MultiPIE

Improves performance on video dataset YouTube Celebrities

Demonstrates robustness to pose variations

Abstract

To address the sequential changes of images including poses, in this paper we propose a recurrent regression neural network(RRNN) framework to unify two classic tasks of cross-pose face recognition on still images and video-based face recognition. To imitate the changes of images, we explicitly construct the potential dependencies of sequential images so as to regularize the final learning model. By performing progressive transforms for sequentially adjacent images, RRNN can adaptively memorize and forget the information that benefits for the final classification. For face recognition of still images, given any one image with any one pose, we recurrently predict the images with its sequential poses to expect to capture some useful information of others poses. For video-based face recognition, the recurrent regression takes one entire sequence rather than one image as its input. We verify RRNN in static face dataset MultiPIE and face video dataset YouTube Celebrities(YTC). The comprehensive experimental results demonstrate the effectiveness of the proposed RRNN method.