AnimeGanV2 Style-Transfer Operator

Authors: filip

Overview

AnimeGanV2 is a style transfer net that transforms images to looking like they fit in an anime movie.

Interface

__init__(self, model_name: str, framework: str = 'pytorch')

Args:

• model_name:
  • which weights to use for inference.
  • supports 'celeba', 'facepaintv1', 'facepaitv2', 'hayao', 'paprika', 'shinkai'
• framework:
  • the framework of the model
  • supported types: str, default is 'pytorch'

__call__(self,  image: 'towhee.types.Image')

Args:

• image:
  • the input image
  • supported types: towhee.types.Image

Returns:

The Operator returns a tuple Tuple[('styled_image', numpy.ndarray)] containing following fields:

• styled_image:
  • styled photo
  • data type: numpy.ndarray
  • shape: (3, x, x)
  • format: RGB
  • values: [0,1]

Requirements

You can get the required python package by requirements.txt.

Reference

Jie Chen, Gang Liu, Xin Chen "AnimeGAN: A Novel Lightweight GAN for Photo Animation." ISICA 2019: Artificial Intelligence Algorithms and Applications pp 242-256, 2019.