CFPL-FAS：用于可泛化人脸反欺骗的无类别提示学习
CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofing
AI总结：
本文提出 CFPL-FAS，一种用于人脸反欺骗的无类别提示学习范式，利用 CQF 和 SQF 分别学习内容和风格提示，并通过提示调制增强模型在未见域上的泛化能力。
摘要：
基于域泛化（DG）的人脸反欺骗（FAS）旨在提升模型在未见域上的性能。现有方法要么依赖域标签来对齐域不变特征空间，要么从整个样本中解耦可泛化特征，这不可避免地导致语义特征结构失真，且泛化能力有限。在本工作中，我们利用如 CLIP 的大规模视觉语言模型（VLM），并借助文本特征动态调整分类器权重，以探索可泛化的视觉特征。具体而言，我们为 DG FAS 提出了一种新颖的无类别提示学习（CFPL）范式，该范式利用两个轻量级 transformer，即 Content Q-Former（CQF）和 Style Q-Former（SQF），分别通过使用一组可学习的查询向量，基于内容和风格特征学习不同的语义提示。因此，可泛化提示可通过两项改进来学习：（1）引入提示文本匹配（PTM）监督，以确保 CQF 学习最能体现内容描述信息的视觉表示。（2）提出多样化风格提示（DSP）技术，通过混合实例特定风格间的特征统计量，使风格提示的学习多样化。最后，所学的文本特征通过设计的提示调制（PM）对视觉特征进行调制以实现泛化。大量实验表明，CFPL 有效性显著，并在多个跨域数据集上优于现有最先进方法。
英文摘要：
Domain generalization (DG) based Face Anti-Spoofing (FAS) aims to improve the model's performance on unseen domains. Existing methods either rely on domain labels to align domain-invariant feature spaces, or disentangle generalizable features from the whole sample, which inevitably lead to the distortion of semantic feature structures and achieve limited generalization. In this work, we make use of large-scale VLMs like CLIP and leverage the textual feature to dynamically adjust the classifier's weights for exploring generalizable visual features. Specifically, we propose a novel Class Free Prompt Learning (CFPL) paradigm for DG FAS, which utilizes two lightweight transformers, namely Content Q-Former (CQF) and Style Q-Former (SQF), to learn the different semantic prompts conditioned on content and style features by using a set of learnable query vectors, respectively. Thus, the generalizable prompt can be learned by two improvements: (1) A Prompt-Text Matched (PTM) supervision is introduced to ensure CQF learns visual representation that is most informative of the content description. (2) A Diversified Style Prompt (DSP) technology is proposed to diversify the learning of style prompts by mixing feature statistics between instance-specific styles. Finally, the learned text features modulate visual features to generalization through the designed Prompt Modulation (PM). Extensive experiments show that the CFPL is effective and outperforms the state-of-the-art methods on several cross-domain datasets.
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。
