Dragness - Pixel2Style2Pixel based face2drag

.gitattributes

@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+shape_predictor_5_face_landmarks.dat filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,8 +1,8 @@
 ---
 title: Dragness
-emoji: 🐢
-colorFrom: green
-colorTo: purple
+emoji: 👸
+colorFrom: yellow
+colorTo: blue
 sdk: gradio
 app_file: app.py
 pinned: false

app.py

@@ -0,0 +1,79 @@
+import os
+#os.system("gdown https://drive.google.com/uc?id=1WEST2O6svlQWpJNomX3947Q2bfJz4bAJ")
+#os.system("gdown https://drive.google.com/uc?id=1CbnhlUI9Tms2o7S2eCg9qwGXZFCyROYy")
+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('dragness_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:
+        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
+
+import gradio as gr
+
+def face2drag(
+    img: Image.Image,
+    size: int
+) -> Image.Image:
+
+    aligned_img = align_first_face(img)
+    input = torch.Tensor(aligned_img)
+    output = net(input)
+    output = tensor2im(output[0])
+    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 = "Dragness"
+description = "Gradio demo for Drag finetuned Pixel2Style2Pixel. To use it, simply upload your image, or click one of the examples to load them. Read more at the links below."
+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='Sample00001.jpg' alt='Sample00001'/><img src='Sample00002.jpg' alt='Sample00002'/><img src='Sample00003.jpg' alt='Sample00003'/><img src='Sample00004.jpg' alt='Sample00004'/><img src='Sample00005.jpg' alt='Sample00005'/><img src='Sample00006.jpg' alt='Sample00006'/></p><p>Drag model was fine tuned by Doron Adler</p>"
+
+examples=[['Example00001.jpg'],['Example00002.jpg'],['Fiona-Example00003.jpg'],['Example00004.jpg'],['Example00005.jpg'],['Example00006.jpg'],['Example00007.jpg']]
+gr.Interface(inference, gr.inputs.Image(type="pil",shape=(256,256)), gr.outputs.Image(type="pil"),title=title,description=description,article=article,examples=examples,enable_queue=True).launch()

dragness_p2s2p_torchscript_cpu.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c758ffe5265d2041a71f369d3eb3565f187aa6bd39f647e1124a66cd79a26f3c
+size 1202678391

face_detection.py

@@ -0,0 +1,140 @@
+# Copyright (c) 2021 Justin Pinkney
+
+import dlib
+import numpy as np
+import os
+from PIL import Image
+from PIL import ImageOps
+from scipy.ndimage import gaussian_filter
+import cv2
+
+
+MODEL_PATH = "shape_predictor_5_face_landmarks.dat"
+detector = dlib.get_frontal_face_detector()
+
+
+def align(image_in, face_index=0, output_size=256):
+    landmarks = list(get_landmarks(image_in))
+    n_faces = len(landmarks)
+    face_index = min(n_faces-1, face_index)
+    if n_faces == 0:
+        aligned_image = image_in
+        quad = None
+    else:
+        aligned_image, quad = image_align(image_in, landmarks[face_index], output_size=output_size)
+
+    return aligned_image, n_faces, quad
+
+
+def composite_images(quad, img, output):
+    """Composite an image into and output canvas according to transformed co-ords"""
+    output = output.convert("RGBA")
+    img = img.convert("RGBA")
+    input_size = img.size
+    src = np.array(((0, 0), (0, input_size[1]), input_size, (input_size[0], 0)), dtype=np.float32)
+    dst = np.float32(quad)
+    mtx = cv2.getPerspectiveTransform(dst, src)
+    img = img.transform(output.size, Image.PERSPECTIVE, mtx.flatten(), Image.BILINEAR)
+    output.alpha_composite(img)
+
+    return output.convert("RGB")
+    
+
+def get_landmarks(image):
+    """Get landmarks from PIL image"""
+    shape_predictor = dlib.shape_predictor(MODEL_PATH)
+
+    max_size = max(image.size)
+    reduction_scale = int(max_size/512)
+    if reduction_scale == 0:
+        reduction_scale = 1
+    downscaled = image.reduce(reduction_scale)
+    img = np.array(downscaled)
+    detections = detector(img, 0)
+    
+    for detection in detections:
+        try:
+            face_landmarks = [(reduction_scale*item.x, reduction_scale*item.y) for item in shape_predictor(img, detection).parts()]
+            yield face_landmarks
+        except Exception as e:
+            print(e)
+
+
+def image_align(src_img, face_landmarks, output_size=512, transform_size=2048, enable_padding=True, x_scale=1, y_scale=1, em_scale=0.1, alpha=False):
+        # Align function modified from ffhq-dataset
+        # See https://github.com/NVlabs/ffhq-dataset for license
+
+        lm = np.array(face_landmarks)
+        lm_eye_left      = lm[2:3]  # left-clockwise
+        lm_eye_right     = lm[0:1]  # left-clockwise
+
+        # Calculate auxiliary vectors.
+        eye_left     = np.mean(lm_eye_left, axis=0)
+        eye_right    = np.mean(lm_eye_right, axis=0)
+        eye_avg      = (eye_left + eye_right) * 0.5
+        eye_to_eye   = 0.71*(eye_right - eye_left)
+        mouth_avg    = lm[4]
+        eye_to_mouth = 1.35*(mouth_avg - eye_avg)
+
+        # Choose oriented crop rectangle.
+        x = eye_to_eye.copy()
+        x /= np.hypot(*x)
+        x *= max(np.hypot(*eye_to_eye) * 2.0, np.hypot(*eye_to_mouth) * 1.8)
+        x *= x_scale
+        y = np.flipud(x) * [-y_scale, y_scale]
+        c = eye_avg + eye_to_mouth * em_scale
+        quad = np.stack([c - x - y, c - x + y, c + x + y, c + x - y])
+        quad_orig = quad.copy()
+        qsize = np.hypot(*x) * 2
+
+        try:
+            src_img = ImageOps.exif_transpose(src_img)
+        except:
+            print("exif problem, not rotating")
+        
+        img = src_img.convert('RGBA').convert('RGB')
+
+        # Shrink.
+        shrink = int(np.floor(qsize / output_size * 0.5))
+        if shrink > 1:
+            rsize = (int(np.rint(float(img.size[0]) / shrink)), int(np.rint(float(img.size[1]) / shrink)))
+            img = img.resize(rsize, Image.ANTIALIAS)
+            quad /= shrink
+            qsize /= shrink
+
+        # Crop.
+        border = max(int(np.rint(qsize * 0.1)), 3)
+        crop = (int(np.floor(min(quad[:,0]))), int(np.floor(min(quad[:,1]))), int(np.ceil(max(quad[:,0]))), int(np.ceil(max(quad[:,1]))))
+        crop = (max(crop[0] - border, 0), max(crop[1] - border, 0), min(crop[2] + border, img.size[0]), min(crop[3] + border, img.size[1]))
+        if crop[2] - crop[0] < img.size[0] or crop[3] - crop[1] < img.size[1]:
+            img = img.crop(crop)
+            quad -= crop[0:2]
+
+        # Pad.
+        pad = (int(np.floor(min(quad[:,0]))), int(np.floor(min(quad[:,1]))), int(np.ceil(max(quad[:,0]))), int(np.ceil(max(quad[:,1]))))
+        pad = (max(-pad[0] + border, 0), max(-pad[1] + border, 0), max(pad[2] - img.size[0] + border, 0), max(pad[3] - img.size[1] + border, 0))
+        if enable_padding and max(pad) > border - 4:
+            pad = np.maximum(pad, int(np.rint(qsize * 0.3)))
+            img = np.pad(np.float32(img), ((pad[1], pad[3]), (pad[0], pad[2]), (0, 0)), 'reflect')
+            h, w, _ = img.shape
+            y, x, _ = np.ogrid[:h, :w, :1]
+            mask = np.maximum(1.0 - np.minimum(np.float32(x) / pad[0], np.float32(w-1-x) / pad[2]), 1.0 - np.minimum(np.float32(y) / pad[1], np.float32(h-1-y) / pad[3]))
+            blur = qsize * 0.02
+            img += (gaussian_filter(img, [blur, blur, 0]) - img) * np.clip(mask * 3.0 + 1.0, 0.0, 1.0)
+            img += (np.median(img, axis=(0,1)) - img) * np.clip(mask, 0.0, 1.0)
+            img = np.uint8(np.clip(np.rint(img), 0, 255))
+            if alpha:
+                mask = 1-np.clip(3.0 * mask, 0.0, 1.0)
+                mask = np.uint8(np.clip(np.rint(mask*255), 0, 255))
+                img = np.concatenate((img, mask), axis=2)
+                img = Image.fromarray(img, 'RGBA')
+            else:
+                img = Image.fromarray(img, 'RGB')
+            quad += pad[:2]
+
+        # Transform.
+        img = img.transform((transform_size, transform_size), Image.QUAD, (quad + 0.5).flatten(), Image.BILINEAR)
+        if output_size < transform_size:
+            img = img.resize((output_size, output_size), Image.ANTIALIAS)
+
+        return img, quad_orig

requirements.txt

@@ -0,0 +1,11 @@
+numpy
+opencv-python
+Pillow
+scikit-image
+torch
+torchvision
+ninja
+dlib
+gdown
+scipy
+cmake

shape_predictor_5_face_landmarks.dat

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4b1e9804792707d3a405c2c16a80a20269e6675021f64a41d30fffafbc41888
+size 9150489