Skip to content

Latest commit

 

History

History
15 lines (11 loc) · 794 Bytes

README.md

File metadata and controls

15 lines (11 loc) · 794 Bytes
Raw

Face-Parsing-Lite - TensorFlow (2.15.0) - Keras

results

Dataset

The masks of CelebAMask-HQ were manually-annotated with the size of 512 x 512 and 19 classes including all facial components and accessories such as skin, nose, eyes, eyebrows, ears, mouth, lip, hair, hat, eyeglass, earring, necklace, neck, and cloth.

Modified Light-Weight Architecture.

  • Invert residual blocks design was used for Squeeze and Exapansion.
  • DepthWise Convolution was used for the balance of parameter efficiency and accuracy.
  • gelu was used in upscale pathway for smooth graident.
  • Trained with a Polynomial Decay Schedule.

Acknowledge

https://github.com/zllrunning/face-parsing.PyTorch