Pose Attention-Guided Profile-to-Frontal Face Recognition

TL;DR

This paper introduces a pose attention-guided network for profile-to-frontal face recognition, leveraging pose as auxiliary information to improve feature extraction and recognition accuracy across various benchmarks.

Contribution

It proposes a novel pose attention block (PAB) and a unified network that enhances profile-to-frontal face recognition by focusing on pose-invariant features.

Findings

Outperforms state-of-the-art methods on Multi-PIE, CFP, IJBC datasets.

Effectively guides feature extraction along channel and spatial dimensions.

Demonstrates robustness in both controlled and in-the-wild scenarios.

Abstract

In recent years, face recognition systems have achieved exceptional success due to promising advances in deep learning architectures. However, they still fail to achieve expected accuracy when matching profile images against a gallery of frontal images. Current approaches either perform pose normalization (i.e., frontalization) or disentangle pose information for face recognition. We instead propose a new approach to utilize pose as an auxiliary information via an attention mechanism. In this paper, we hypothesize that pose attended information using an attention mechanism can guide contextual and distinctive feature extraction from profile faces, which further benefits a better representation learning in an embedded domain. To achieve this, first, we design a unified coupled profile-to-frontal face recognition network. It learns the mapping from faces to a compact embedding subspace via a class-specific contrastive loss. Second, we develop a novel pose attention block (PAB) to specially guide the pose-agnostic feature extraction from profile faces. To be more specific, PAB is designed to explicitly help the network to focus on important features along both channel and spatial dimension while learning discriminative yet pose invariant features in an embedding subspace. To validate the effectiveness of our proposed method, we conduct experiments on both controlled and in the wild benchmarks including Multi-PIE, CFP, IJBC, and show superiority over the state of the arts.