Search papers, labs, and topics across Lattice.
DCMorph, a novel face morphing attack framework, uses a dual-stream diffusion approach operating on both identity conditioning and latent spaces to generate high-fidelity morphs. It employs decoupled cross-attention interpolation to inject identity features from both source faces during denoising and uses DDIM inversion with spherical latent interpolation for geometric consistency. Experiments on four face recognition systems show DCMorph achieves superior attack success rates compared to existing methods, while also evading current morphing attack detection.
Face recognition systems, beware: DCMorph's dual-stream diffusion morphs achieve unprecedented attack success rates while remaining stealthy to existing detection methods.
Advancing face morphing attack techniques is crucial to anticipate evolving threats and develop robust defensive mechanisms for identity verification systems. This work introduces DCMorph, a dual-stream diffusion-based morphing framework that simultaneously operates at both identity conditioning and latent space levels. Unlike image-level methods suffering from blending artifacts or GAN-based approaches with limited reconstruction fidelity, DCMorph leverages identity-conditioned latent diffusion models through two mechanisms: (1) decoupled cross-attention interpolation that injects identity-specific features from both source faces into the denoising process, enabling explicit dual-identity conditioning absent in existing diffusion-based methods, and (2) DDIM inversion with spherical interpolation between inverted latent representations from both source faces, providing geometrically consistent initial latent representation that preserves structural attributes. Vulnerability analyses across four state-of-the-art face recognition systems demonstrate that DCMorph achieves the highest attack success rates compared to existing methods at both operational thresholds, while remaining challenging to detect by current morphing attack detection solutions.
