InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing

TL;DR

InstructFLIP is a novel instruction-tuned vision-language framework for face anti-spoofing that improves cross-domain generalization by leveraging textual guidance and decoupling content and style instructions, reducing training redundancy.

Contribution

The paper introduces InstructFLIP, a unified vision-language model that enhances face anti-spoofing by integrating textual instructions and a meta-domain strategy for better generalization.

Findings

Outperforms state-of-the-art models in accuracy.

Reduces training redundancy across multiple domains.

Effectively decouples content and style instructions.

Abstract

Face anti-spoofing (FAS) aims to construct a robust system that can withstand diverse attacks. While recent efforts have concentrated mainly on cross-domain generalization, two significant challenges persist: limited semantic understanding of attack types and training redundancy across domains. We address the first by integrating vision-language models (VLMs) to enhance the perception of visual input. For the second challenge, we employ a meta-domain strategy to learn a unified model that generalizes well across multiple domains. Our proposed InstructFLIP is a novel instruction-tuned framework that leverages VLMs to enhance generalization via textual guidance trained solely on a single domain. At its core, InstructFLIP explicitly decouples instructions into content and style components, where content-based instructions focus on the essential semantics of spoofing, and style-based instructions consider variations related to the environment and camera characteristics. Extensive experiments demonstrate the effectiveness of InstructFLIP by outperforming SOTA models in accuracy and substantially reducing training redundancy across diverse domains in FAS. Project website is available at https://kunkunlin1221.github.io/InstructFLIP.