FaceCoT: A Benchmark Dataset for Face Anti-Spoofing with Chain-of-Thought Reasoning
Journal:
arXiv
Published Date:
Jun 2, 2025
Abstract
Face Anti-Spoofing (FAS) typically depends on a single visual modality when
defending against presentation attacks such as print attacks, screen replays,
and 3D masks, resulting in limited generalization across devices, environments,
and attack types. Meanwhile, Multimodal Large Language Models (MLLMs) have
recently achieved breakthroughs in image-text understanding and semantic
reasoning, suggesting that integrating visual and linguistic co-inference into
FAS can substantially improve both robustness and interpretability. However,
the lack of a high-quality vision-language multimodal dataset has been a
critical bottleneck. To address this, we introduce FaceCoT (Face
Chain-of-Thought), the first large-scale Visual Question Answering (VQA)
dataset tailored for FAS. FaceCoT covers 14 spoofing attack types and enriches
model learning with high-quality CoT VQA annotations. Meanwhile, we develop a
caption model refined via reinforcement learning to expand the dataset and
enhance annotation quality. Furthermore, we introduce a CoT-Enhanced
Progressive Learning (CEPL) strategy to better leverage the CoT data and boost
model performance on FAS tasks. Extensive experiments demonstrate that models
trained with FaceCoT and CEPL outperform state-of-the-art methods on multiple
benchmark datasets.
