Expression-Diverse References for Identity-Preserving Video Generation

Key points

• First, we quantify how face-recognition similarity varies with expression intensity using controlled photographs and MEAD videos.1
• We then construct a compact yet expressive reference gallery that captures diverse expression-dependent facial configurations.1
• Matching against this gallery provides a more robust measure of identity similarity under expressive motion.1
• To further expose performance degradation with expression intensity, we report identity similarity separately for mild, intense, and extreme expressions.1

Sources (1)

• [1]Expression-Diverse References for Identity-Preserving Video Generation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 12:15 AM

  Identity-preserving video generation aims to maintain a subject's identity while synthesizing realistic videos.

  First, we quantify how face-recognition similarity varies with expression intensity using controlled photographs and MEAD videos.

Extractive summary: sentences quoted from the sources.