Backlog

Features

  • Add GIF support
  • Feature manipulation using interfacegan or styleganx
  • Add the face enhancer gpen_bfr_1024 and gpen_bfr_2048 model
  • Introduce frame colorizer using BigColor or similar
  • Detect the optimal configuration with the benchmark suite
  • Frame interpolation like Super-SloMo
  • Support DeepFaceLab models
  • Extend face enhancer with skin restoration model
  • Support for 180 and 360 degree videos
  • Provide selection from multi-source variants
  • Ignore sources without a face in a multi-source context
  • Introduce a multi inference pool to support multiple GPUs
  • Replace gender age models by non-commercial alternative
  • Validate defaults from facefusion.ini file

Bugs

  • Fix retinaface not being able to detect cropped face
  • Fix preview of lip syncer when using --trim-frame-start
Last modified 3d ago