File size: 2,222 Bytes
c7f097c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
04a4626
 
 
c7f097c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# from https://huggingface.co/spaces/eugenesiow/remove-bg/blob/main/app.py
import cv2
import torch
import numpy as np
from torchvision import transforms

class RemoveBackground(object):
    def __init__(self):
        self.model = torch.hub.load('pytorch/vision:v0.6.0', 'deeplabv3_resnet101', pretrained=True)
        self.model.eval()

    def make_transparent_foreground(self, pic, mask):
        # split the image into channels
        b, g, r = cv2.split(np.array(pic).astype('uint8'))
        # add an alpha channel with and fill all with transparent pixels (max 255)
        a = np.ones(mask.shape, dtype='uint8') * 255
        # merge the alpha channel back
        alpha_im = cv2.merge([b, g, r, a], 4)
        # create a transparent background
        bg = np.zeros(alpha_im.shape)
        # setup the new mask
        new_mask = np.stack([mask, mask, mask, mask], axis=2)
        # copy only the foreground color pixels from the original image where mask is set
        foreground = np.where(new_mask, alpha_im, bg).astype(np.uint8)

        return foreground


    def remove_background(self, input_image):
        preprocess = transforms.Compose([
            transforms.ToTensor(),
            transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
        ])

        input_tensor = preprocess(input_image)
        input_batch = input_tensor.unsqueeze(0)  # create a mini-batch as expected by the model

        # move the input and model to GPU for speed if available
        if torch.cuda.is_available():
           input_batch = input_batch.to('cuda')
           self.model.to('cuda')

        with torch.no_grad():
            output = self.model(input_batch)['out'][0]
        output_predictions = output.argmax(0)

        # create a binary (black and white) mask of the profile foreground
        mask = output_predictions.byte().cpu().numpy()
        background = np.zeros(mask.shape)
        bin_mask = np.where(mask, 255, background).astype(np.uint8)

        foreground = self.make_transparent_foreground(input_image, bin_mask)

        return foreground, bin_mask

    def inference(self, img):
        foreground, _ = self.remove_background(img)
        return foreground