arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.03037cs.CV
收藏

CATFace:用于低质量人脸识别的基于自注意力蒸馏的跨属性引导 Transformer

CATFace: Cross-Attribute-Guided Transformer with Self-Attention Distillation for Low-Quality Face Recognition

Niloufar Alipour Talemi, Hossein Kashiani, Nasser M. Nasrabadi

AI总结:

本文提出一种多分支神经网络 CATFace,利用软生物特征属性信息,通过跨属性引导 Transformer 融合模块和自注意力蒸馏框架提升低质量人脸识别性能。

摘要:

尽管人脸识别(FR)在近年来取得了巨大成功,但由于面部细节模糊,在低质量图像中准确识别人脸仍然具有挑战性。然而,在处理低质量图像时,预测特定的软生物特征(SB)属性(如性别和秃发)通常是可行的。在本文中,我们提出了一种新颖的多分支神经网络,利用 SB 属性信息来提升 FR 的性能。为此,我们提出了跨属性引导 Transformer 融合(CATF)模块,该模块能有效捕捉 FR 和 SB 特征表示之间的长程依赖和关系。所提出的 CATF 模块在双重交叉注意力操作中通过信息的双向流动产生的协同作用增强了 FR 的性能。此外,我们引入了一种新颖的自注意力蒸馏框架,通过在特征空间中将低质量图像与其高质量对应图像对齐,有效地突出关键面部区域(如面部特征点)。所提出的自注意力蒸馏正则化我们的网络,使其在无约束环境中学习统一的质量不变特征表示。我们在质量各异的各种 FR 基准上进行了大量实验。实验结果表明,与最先进的 FR 研究相比,我们的 FR 方法具有优越性。

英文摘要:

Although face recognition (FR) has achieved great success in recent years, it is still challenging to accurately recognize faces in low-quality images due to the obscured facial details. Nevertheless, it is often feasible to make predictions about specific soft biometric (SB) attributes, such as gender, and baldness even in dealing with low-quality images. In this paper, we propose a novel multi-branch neural network that leverages SB attribute information to boost the performance of FR. To this end, we propose a cross-attribute-guided transformer fusion (CATF) module that effectively captures the long-range dependencies and relationships between FR and SB feature representations. The synergy created by the reciprocal flow of information in the dual cross-attention operations of the proposed CATF module enhances the performance of FR. Furthermore, we introduce a novel self-attention distillation framework that effectively highlights crucial facial regions, such as landmarks by aligning low-quality images with those of their high-quality counterparts in the feature space. The proposed self-attention distillation regularizes our network to learn a unified quality-invariant feature representation in unconstrained environments. We conduct extensive experiments on various FR benchmarks varying in quality. Experimental results demonstrate the superiority of our FR method compared to state-of-the-art FR studies.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑