app.py

import gradio as gr
import torch
from torch import nn
import imageio
import cv2


class Generator(nn.Module):
    # Refer to the link below for explanations about nc, nz, and ngf
    # https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html#inputs
    def __init__(self, nc=4, nz=100, ngf=64):
        super(Generator, self).__init__()
        self.network = nn.Sequential(
            nn.ConvTranspose2d(nz, ngf * 4, 3, 1, 0, bias=False),
            nn.BatchNorm2d(ngf * 4),
            nn.ReLU(True),
            nn.ConvTranspose2d(ngf * 4, ngf * 2, 3, 2, 1, bias=False),
            nn.BatchNorm2d(ngf * 2),
            nn.ReLU(True),
            nn.ConvTranspose2d(ngf * 2, ngf, 4, 2, 0, bias=False),
            nn.BatchNorm2d(ngf),
            nn.ReLU(True),
            nn.ConvTranspose2d(ngf, nc, 4, 2, 1, bias=False),
            nn.Tanh(),
        )

    def forward(self, input):
        output = self.network(input)
        return output


def display_gif(file_name, save_name):
    images = []

    for frame in range(8):
        frame_name = '%d' % (frame)
        image_filename = file_name + frame_name + '.png'
        images.append(imageio.imread(image_filename))

    gif_filename = 'avatar_source.gif'
    return imageio.mimsave(gif_filename, images)


def display_gif_pad(file_name, save_name):
    images = []

    for frame in range(8):
        frame_name = '%d' % (frame)
        image_filename = file_name + frame_name + '.png'
        image = imageio.imread(image_filename)
        image = image[:, :, :3]
        image_pad = cv2.copyMakeBorder(image, 0, 0, 100, 100, cv2.BORDER_CONSTANT, value=0)
        images.append(image_pad)

    gif_filename = 'avatar_source.gif'
    return imageio.mimsave(gif_filename, images)
    

def display_image(file_name):

    image_filename = file_name + '0' + '.png'
    print(image_filename)
    image = imageio.imread(image_filename)
    imageio.imwrite('image.png', image)
    

def run(action, hair, top, bottom):

    # == Source Avatar ==
    # body
    #if body == "human":body = '0'
    #elif body == "alien": body = '1'
    body = '0'
    
    # hair
    if hair == "green": hair = '0'
    elif hair == "yellow": hair = '2'
    elif hair == "rose": hair = '4'
    elif hair == "red": hair = '7'
    elif hair == "wine": hair = '8'
    
    # top
    if top == "brown": top = '0'
    elif top == "blue": top = '1'
    elif top == "white": top = '2'
    
    # bottom
    if bottom == "white": bottom = '0'
    elif bottom == "golden": bottom = '1'
    elif bottom == "red": bottom = '2'
    elif bottom == "silver": bottom = '3'
    
    file_name_source = './Sprite/frames/domain_1/' + action + '/'
    file_name_source = file_name_source + 'front' + '_' + str(body) + str(bottom) + str(top) + str(hair) + '_'
    
    gif = display_gif_pad(file_name_source, 'avatar_source.gif')
    
    # == Target Avatar ==
    body_target = '1'
    hair_target = np.random.choice('1', '3', '5', '6', '9')
    top_target = np.random.choice('3', '4', '5', '6')
    bottom_target = np.random.choice('4', '5', '6')
    
    file_name_target = './Sprite/frames/domain_2/' + action + '/'
    file_name_target = file_name_target + 'front' + '_' + str(body_target) + str(bottom_target) + str(top_target) + str(hair_target) + '_'
    
    gif = display_gif_pad(file_name_target, 'avatar_target.gif')
    
    return 'avatar_source.gif', 'avatar_target.gif'


gr.Interface(
    run,
    inputs=[
        gr.Radio(choices=["shoot", "slash", "spellcard", "thrust", "walk"], value="shoot"),
        gr.Radio(choices=["green", "yellow", "rose", "red", "wine"], value="green"),
        gr.Radio(choices=["brown", "blue", "white"], value="brown"),
        gr.Radio(choices=["white", "golden", "red", "silver"], value="white"),
    ],
    outputs=[
        gr.components.Image(type="file", label="Source Avatar (Costumed by You)"),
        gr.components.Image(type="file", label="Target Avatar (Randomly Chosen)"),
    ],
    live=True,
    title="TransferVAE for Unsupervised Video Domain Adaptation",
).launch()