We provide two pre-trained models: human to anime and human to cats
Run the following command to translate the demo inputs. Note that this command works the best for models trained using twingan. You'll have to change this a little to make it work with image_generation models.
# Make sure you're under
python inference/image_translation_infer.py \
--model_path="/PATH/TO/MODEL/"
--image_hw=256 # 256 for anime, 128 for cats.
--input_tensor_name="sources_ph"
--output_tensor_name="custom_generated_t_style_source:0"
--input_image_path="./demo/inference_input/cropped"
--output_image_path="./demo/inference_output/temp"