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/FSA-Net (Caps-Fusion)

FSA-Net (Caps-Fusion)

Reported on 18 benchmarks across 3 tasks

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

Computer Vision6 results

  • Pose EstimationonAFLW2000
    Geodesic Error (GE)
    8.16
    best: 5.23 (OpNet)
  • Pose EstimationonAFLW2000
    MAE
    5.07
    best: 3.15 (OpNet)
  • Pose EstimationonBIWI
    Geodesic Error (GE)
    7.64
    best: 7.01 (OpNet)
  • Pose EstimationonBIWI
    Geodesic Error - aligned (GE)
    5.36
    best: 4.72 (OpNet)
  • Pose EstimationonBIWI
    MAE (trained with other data)
    4
    best: 2.75 (TRG (w/ 300WLP))
  • Pose EstimationonBIWI
    MAE-aligned (trained with other data)
    2.92
    best: 2.65 (OpNet)

Methodology6 results

  • 3DonAFLW2000
    Geodesic Error (GE)
    8.16
    best: 5.23 (OpNet)
  • 3DonAFLW2000
    MAE
    5.07
    best: 3.15 (OpNet)
  • 3DonBIWI
    Geodesic Error (GE)
    7.64
    best: 7.01 (OpNet)
  • 3DonBIWI
    Geodesic Error - aligned (GE)
    5.36
    best: 4.72 (OpNet)
  • 3DonBIWI
    MAE (trained with other data)
    4
    best: 2.75 (TRG (w/ 300WLP))
  • 3DonBIWI
    MAE-aligned (trained with other data)
    2.92
    best: 2.65 (OpNet)

Audio6 results

  • 1 Image, 2*2 StitchionAFLW2000
    Geodesic Error (GE)
    8.16
    best: 5.23 (OpNet)
  • 1 Image, 2*2 StitchionAFLW2000
    MAE
    5.07
    best: 3.15 (OpNet)
  • 1 Image, 2*2 StitchionBIWI
    Geodesic Error (GE)
    7.64
    best: 7.01 (OpNet)
  • 1 Image, 2*2 StitchionBIWI
    Geodesic Error - aligned (GE)
    5.36
    best: 4.72 (OpNet)
  • 1 Image, 2*2 StitchionBIWI
    MAE (trained with other data)
    4
    best: 2.75 (TRG (w/ 300WLP))
  • 1 Image, 2*2 StitchionBIWI
    MAE-aligned (trained with other data)
    2.92
    best: 2.65 (OpNet)