File size: 12,926 Bytes
8ed2f16
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
#################################################
# Copyright (c) 2021-present, xiaobing.ai, Inc. #
# All rights reserved.                          #
#################################################
# CV Research, DEV(USA) xiaobing.               #
# written by [email protected]             #
#################################################

##### python internal and external package
import os
import cv2
import torch
import torch.nn as nn
import numpy as np
import torchvision.transforms as transforms
from PIL import Image
import math

##### self defined package
from lib.models.ldmk.hrnet import LandmarkDetector

gauss_kernel = None

class ldmkDetector(nn.Module):
    def __init__(self, cfg):
        super(ldmkDetector, self).__init__()
        if cfg.model.ldmk.model_name == "h3r":
            self.model = LandmarkDetector(cfg.model.ldmk.model_path)
        else:
            print("Error: the model {} of landmark is not exists".format(cfg.model.ldmk.model_name))

        self.model.eval()
        self.model.cuda()
        
        self.size = cfg.model.ldmk.img_size # 256
        
        self.landmark_transform = transforms.Compose([
            transforms.Resize(size=(self.size, self.size)),
            transforms.ToTensor(),
            transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
        ])
        
    def _transform(self, img):
        
        h, w, c = img.shape
        img = img[:, :, ::-1]
        img = Image.fromarray(img.astype(np.uint8))
        img = self.landmark_transform(img)
        img = img.type(torch.FloatTensor).unsqueeze(0)
        
        
        img = img.cuda()
        
        return img, h, w
        
    def forward(self, img):
        img, h, w  = self._transform(img)
        
        _, landmarks = self.model(img)
        
        landmarks = landmarks / torch.Tensor([self.size / w, self.size / h]).reshape(1, 1, 2).cuda()
        
        landmarks = landmarks.detach().cpu().numpy()
        
        return landmarks


class ldmk3dDetector(nn.Module):
    def __init__(self, cfg):
        super(ldmk3dDetector, self).__init__()
        self.model_3d = torch.jit.load(cfg.model.ldmk_3d.model_path)
        self.model_depth = torch.jit.load(cfg.model.ldmk_3d.model_depth_path)

        self.model_3d.eval()
        self.model_depth.eval()
        self.model_3d.cuda()
        self.model_depth.cuda()
        
        self.size = cfg.model.ldmk.img_size # 256
        
        self.landmark_transform = transforms.Compose([
            transforms.Resize(size=(self.size, self.size)),
            transforms.ToTensor(),
        ])
        
    def _transform(self, img):
        
        h, w, c = img.shape
        img = img[:, :, ::-1]
        img = Image.fromarray(img.astype(np.uint8))
        img = self.landmark_transform(img)
        img = img.type(torch.FloatTensor).unsqueeze(0)
        
        img = img.cuda()
        
        return img, h, w

    def get_cropped_img(self, img, box):
        center = torch.tensor(
            [box[2] - (box[2] - box[0]) / 2.0, box[3] - (box[3] - box[1]) / 2.0])
        center[1] = center[1] - (box[3] - box[1]) * 0.12
        scale = (box[2] - box[0] + box[3] - box[1]) / 192

        inp = crop(img, center, scale)
        return inp, center, scale
        
    def forward(self, img, boxes):
        ldmks = []
        for box in boxes:
            img_cropped, center, scale = self.get_cropped_img(img, box)
            img_cropped, h, w  = self._transform(img_cropped)
            
            out = self.model_3d(img_cropped).detach()
            out = out.cpu().numpy()
            
            pts, pts_img, scores = get_preds_fromhm(out, center.numpy(), scale)
            pts, pts_img = torch.from_numpy(pts), torch.from_numpy(pts_img)
            pts, pts_img = pts.view(68, 2) * 4, pts_img.view(68, 2)
            scores = scores.squeeze(0)

            heatmaps = np.zeros((68, 256, 256), dtype=np.float32)
            for i in range(68):
                if pts[i, 0] > 0 and pts[i, 1] > 0:
                    heatmaps[i] = draw_gaussian(
                        heatmaps[i], pts[i], 2)
            heatmaps = torch.from_numpy(
                heatmaps).unsqueeze_(0)

            heatmaps = heatmaps.cuda()
            depth_pred = self.model_depth(
                torch.cat((img_cropped, heatmaps), 1)).data.cpu().view(68, 1)
            pts_img = torch.cat(
                (pts_img, depth_pred * (1.0 / (256.0 / (200.0 * scale)))), 1).detach().cpu().numpy()
            
            ldmks.append(pts_img)
        return np.array(ldmks)

def get_preds_fromhm(hm, center=None, scale=None):
    """Obtain (x,y) coordinates given a set of N heatmaps. If the center
    and the scale is provided the function will return the points also in
    the original coordinate frame.
    Arguments:
        hm {torch.tensor} -- the predicted heatmaps, of shape [B, N, W, H]
    Keyword Arguments:
        center {torch.tensor} -- the center of the bounding box (default: {None})
        scale {float} -- face scale (default: {None})
    """
    B, C, H, W = hm.shape
    hm_reshape = hm.reshape(B, C, H * W)
    idx = np.argmax(hm_reshape, axis=-1)
    scores = np.take_along_axis(hm_reshape, np.expand_dims(idx, axis=-1), axis=-1).squeeze(-1)
    preds, preds_orig = _get_preds_fromhm(hm, idx, center, scale)

    return preds, preds_orig, scores

def _get_preds_fromhm(hm, idx, center=None, scale=None):
    """Obtain (x,y) coordinates given a set of N heatmaps and the
    coresponding locations of the maximums. If the center
    and the scale is provided the function will return the points also in
    the original coordinate frame.
    Arguments:
        hm {torch.tensor} -- the predicted heatmaps, of shape [B, N, W, H]
    Keyword Arguments:
        center {torch.tensor} -- the center of the bounding box (default: {None})
        scale {float} -- face scale (default: {None})
    """
    B, C, H, W = hm.shape
    idx += 1
    preds = idx.repeat(2).reshape(B, C, 2).astype(np.float32)
    preds[:, :, 0] = (preds[:, :, 0] - 1) % W + 1
    preds[:, :, 1] = np.floor((preds[:, :, 1] - 1) / H) + 1

    for i in range(B):
        for j in range(C):
            hm_ = hm[i, j, :]
            pX, pY = int(preds[i, j, 0]) - 1, int(preds[i, j, 1]) - 1
            if pX > 0 and pX < 63 and pY > 0 and pY < 63:
                diff = np.array(
                    [hm_[pY, pX + 1] - hm_[pY, pX - 1],
                     hm_[pY + 1, pX] - hm_[pY - 1, pX]])
                preds[i, j] += np.sign(diff) * 0.25

    preds -= 0.5

    preds_orig = np.zeros_like(preds)
    if center is not None and scale is not None:
        for i in range(B):
            for j in range(C):
                preds_orig[i, j] = transform_np(
                    preds[i, j], center, scale, H, True)

    return preds, preds_orig

def draw_gaussian(image, point, sigma):
    global gauss_kernel
    # Check if the gaussian is inside
    ul = [math.floor(point[0] - 3 * sigma), math.floor(point[1] - 3 * sigma)]
    br = [math.floor(point[0] + 3 * sigma), math.floor(point[1] + 3 * sigma)]
    if (ul[0] > image.shape[1] or ul[1] > image.shape[0] or br[0] < 1 or br[1] < 1):
        return image
    size = 6 * sigma + 1
    if gauss_kernel is None:
        g = _gaussian(size)
        gauss_kernel = g
    else:
        g = gauss_kernel
    g_x = [int(max(1, -ul[0])), int(min(br[0], image.shape[1])) - int(max(1, ul[0])) + int(max(1, -ul[0]))]
    g_y = [int(max(1, -ul[1])), int(min(br[1], image.shape[0])) - int(max(1, ul[1])) + int(max(1, -ul[1]))]
    img_x = [int(max(1, ul[0])), int(min(br[0], image.shape[1]))]
    img_y = [int(max(1, ul[1])), int(min(br[1], image.shape[0]))]
    assert (g_x[0] > 0 and g_y[1] > 0)
    image[img_y[0] - 1:img_y[1], img_x[0] - 1:img_x[1]
          ] = image[img_y[0] - 1:img_y[1], img_x[0] - 1:img_x[1]] + g[g_y[0] - 1:g_y[1], g_x[0] - 1:g_x[1]]
    image[image > 1] = 1
    return image

def crop(image, center, scale, resolution=256.0):
    """Center crops an image or set of heatmaps
    Arguments:
        image {numpy.array} -- an rgb image
        center {numpy.array} -- the center of the object, usually the same as of the bounding box
        scale {float} -- scale of the face
    Keyword Arguments:
        resolution {float} -- the size of the output cropped image (default: {256.0})
    Returns:
        [type] -- [description]
    """  # Crop around the center point
    """ Crops the image around the center. Input is expected to be an np.ndarray """
    ul = transform([1, 1], center, scale, resolution, True)
    br = transform([resolution, resolution], center, scale, resolution, True)
    # pad = math.ceil(torch.norm((ul - br).float()) / 2.0 - (br[0] - ul[0]) / 2.0)
    if image.ndim > 2:
        newDim = np.array([br[1] - ul[1], br[0] - ul[0],
                           image.shape[2]], dtype=np.int32)
        newImg = np.zeros(newDim, dtype=np.uint8)
    else:
        newDim = np.array([br[1] - ul[1], br[0] - ul[0]], dtype=np.int)
        newImg = np.zeros(newDim, dtype=np.uint8)
    ht = image.shape[0]
    wd = image.shape[1]
    newX = np.array(
        [max(1, -ul[0] + 1), min(br[0], wd) - ul[0]], dtype=np.int32)
    newY = np.array(
        [max(1, -ul[1] + 1), min(br[1], ht) - ul[1]], dtype=np.int32)
    oldX = np.array([max(1, ul[0] + 1), min(br[0], wd)], dtype=np.int32)
    oldY = np.array([max(1, ul[1] + 1), min(br[1], ht)], dtype=np.int32)
    newImg[newY[0] - 1:newY[1], newX[0] - 1:newX[1]
           ] = image[oldY[0] - 1:oldY[1], oldX[0] - 1:oldX[1], :]
    newImg = cv2.resize(newImg, dsize=(int(resolution), int(resolution)),
                        interpolation=cv2.INTER_LINEAR)
    return newImg

def transform(point, center, scale, resolution, invert=False):
    """Generate and affine transformation matrix.
    Given a set of points, a center, a scale and a targer resolution, the
    function generates and affine transformation matrix. If invert is ``True``
    it will produce the inverse transformation.
    Arguments:
        point {torch.tensor} -- the input 2D point
        center {torch.tensor or numpy.array} -- the center around which to perform the transformations
        scale {float} -- the scale of the face/object
        resolution {float} -- the output resolution
    Keyword Arguments:
        invert {bool} -- define wherever the function should produce the direct or the
        inverse transformation matrix (default: {False})
    """
    _pt = torch.ones(3)
    _pt[0] = point[0]
    _pt[1] = point[1]

    h = 200.0 * scale
    t = torch.eye(3)
    t[0, 0] = resolution / h
    t[1, 1] = resolution / h
    t[0, 2] = resolution * (-center[0] / h + 0.5)
    t[1, 2] = resolution * (-center[1] / h + 0.5)

    if invert:
        t = torch.inverse(t)

    new_point = (torch.matmul(t, _pt))[0:2]

    return new_point.int()

def transform_np(point, center, scale, resolution, invert=False):
    """Generate and affine transformation matrix.
    Given a set of points, a center, a scale and a targer resolution, the
    function generates and affine transformation matrix. If invert is ``True``
    it will produce the inverse transformation.
    Arguments:
        point {numpy.array} -- the input 2D point
        center {numpy.array} -- the center around which to perform the transformations
        scale {float} -- the scale of the face/object
        resolution {float} -- the output resolution
    Keyword Arguments:
        invert {bool} -- define wherever the function should produce the direct or the
        inverse transformation matrix (default: {False})
    """
    _pt = np.ones(3)
    _pt[0] = point[0]
    _pt[1] = point[1]

    h = 200.0 * scale
    t = np.eye(3)
    t[0, 0] = resolution / h
    t[1, 1] = resolution / h
    t[0, 2] = resolution * (-center[0] / h + 0.5)
    t[1, 2] = resolution * (-center[1] / h + 0.5)

    if invert:
        t = np.ascontiguousarray(np.linalg.pinv(t))

    new_point = np.dot(t, _pt)[0:2]

    return new_point.astype(np.int32)

def _gaussian(
        size=3, sigma=0.25, amplitude=1, normalize=False, width=None,
        height=None, sigma_horz=None, sigma_vert=None, mean_horz=0.5,
        mean_vert=0.5):
    # handle some defaults
    if width is None:
        width = size
    if height is None:
        height = size
    if sigma_horz is None:
        sigma_horz = sigma
    if sigma_vert is None:
        sigma_vert = sigma
    center_x = mean_horz * width + 0.5
    center_y = mean_vert * height + 0.5
    gauss = np.empty((height, width), dtype=np.float32)
    # generate kernel
    for i in range(height):
        for j in range(width):
            gauss[i][j] = amplitude * math.exp(-(math.pow((j + 1 - center_x) / (
                sigma_horz * width), 2) / 2.0 + math.pow((i + 1 - center_y) / (sigma_vert * height), 2) / 2.0))
    if normalize:
        gauss = gauss / np.sum(gauss)
    return gauss