app.py

import os
import sys
import face_detection
import PIL
from PIL import Image, ImageOps
import numpy as np

import torch
torch.set_grad_enabled(False)
net = torch.jit.load('apocalyptify_p2s2p_torchscript_cpu.pt')
net.eval()


def tensor2im(var):
	var = var.cpu().detach().transpose(0, 2).transpose(0, 1).numpy()
	var = ((var + 1) / 2)
	var[var < 0] = 0
	var[var > 1] = 1
	var = var * 255
	return Image.fromarray(var.astype('uint8'))

def image_as_array(image_in):
    im_array = np.array(image_in, np.float32)
    im_array = (im_array/255)*2 - 1
    im_array = np.transpose(im_array, (2, 0, 1))
    im_array = np.expand_dims(im_array, 0)
    return im_array

def find_aligned_face(image_in, size=256):   
    aligned_image, n_faces, quad = face_detection.align(image_in, face_index=0, output_size=size)
    return aligned_image, n_faces, quad

def align_first_face(image_in, size=256):
    aligned_image, n_faces, quad = find_aligned_face(image_in,size=size)
    if n_faces == 0:
        try:
            image_in = ImageOps.exif_transpose(image_in)
        except:
            print("exif problem, not rotating")
        image_in = image_in.resize((size, size))
        im_array = image_as_array(image_in)
    else:
        im_array = image_as_array(aligned_image)

    return im_array

def img_concat_h(im1, im2):
    dst = Image.new('RGB', (im1.width + im2.width, im1.height))
    dst.paste(im1, (0, 0))
    dst.paste(im2, (im1.width, 0))
    return dst

import gradio as gr

def face2drag(
    img: Image.Image,
    size: int
) -> Image.Image:

    aligned_img = align_first_face(img)
    if aligned_img is None:
        output=None
    else:
        input = torch.Tensor(aligned_img)
        output = net(input)
        output = tensor2im(output[0])
        output = img_concat_h(tensor2im(torch.Tensor(aligned_img)[0]), output)

    return output
    
import os
import collections
from typing import Union, List
import numpy as np
from PIL import Image
import PIL.Image
import PIL.ImageFile
import numpy as np
import scipy.ndimage
import requests

def inference(img):
    out = face2drag(img, 256)
    return out
      
  
title = "Apocalyptify"
description = "How will your face look after the Apocalypse? Will you become a Zombie? A Ghoul? A person who fights them? Upload an image with a face, or click one of the examples below. If a face could not be detected, A face will still be created based on the features of the input."
article = "<p style='text-align: center'><a href='https://github.com/justinpinkney/pixel2style2pixel/tree/nw' target='_blank'>Github Repo</a></p><p style='text-align: center'>samples: <img src='https://hf.space/gradioiframe/Norod78/Apocalyptify/file/Sample00001.jpg' alt='Sample00001'/><img src='https://hf.space/gradioiframe/Norod78/Apocalyptify/file/Sample00002.jpg' alt='Sample00002'/><img src='https://hf.space/gradioiframe/Norod78/Apocalyptify/file/Sample00003.jpg' alt='Sample00003'/><img src='https://hf.space/gradioiframe/Norod78/Apocalyptify/file/Sample00004.jpg' alt='Sample00004'/><img src='https://hf.space/gradioiframe/Norod78/Apocalyptify/file/Sample00005.jpg' alt='Sample00005'/></p><p>The Apocalypse model was fine tuned on a pre-trained Pixel2Style2Pixel model by Doron Adler</p>"

examples=[['Example00001.jpg'],['Example00002.jpg'],['Example00003.jpg'],['Example00004.jpg'],['Example00005.jpg'], ['Example00006.jpg']]

demo = gr.Interface(
    inference, 
    inputs=[gr.Image(type="pil", label="Input")],
    outputs=[gr.Image(type="pil", label="Output")],
    title=title,
    description=description,
    article=article,
    examples=examples,
    allow_flagging="never"
    )

demo.queue()
demo.launch()