TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/X2Face

X2Face

Reported on 27 benchmarks across 9 tasks

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision12 results

  • Talking Head GenerationonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • Talking Head GenerationonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • Talking Head GenerationonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)
  • Face GenerationonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • Face GenerationonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • Face GenerationonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)
  • Face ReconstructiononVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • Face ReconstructiononVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • Face ReconstructiononVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)
  • 3D Face ReconstructiononVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • 3D Face ReconstructiononVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • 3D Face ReconstructiononVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)

Medical6 results

  • Image GenerationonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • Image GenerationonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • Image GenerationonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)
  • 3D Face ModellingonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • 3D Face ModellingonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • 3D Face ModellingonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)

Music3 results

  • Facial Recognition and ModellingonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • Facial Recognition and ModellingonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • Facial Recognition and ModellingonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)

Methodology3 results

  • 3DonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • 3DonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • 3DonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)

Audio3 results

  • 10-shot image generationonVoxCeleb1 - 8-shot learning
    FID
    51.5
    best: 38 (Few-shot Adversarial Model)
  • 10-shot image generationonVoxCeleb1 - 32-shot learning
    FID
    56.5
    best: 29.5 (Few-shot Adversarial Model)
  • 10-shot image generationonVoxCeleb1 - 1-shot learning
    FID
    45.8
    best: 43 (Few-shot Adversarial Model)