File size: 5,711 Bytes
689e965
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import gradio as gr

import cv2

import torch

import matplotlib.pyplot as plt
from matplotlib import cm
from matplotlib import colors
from mpl_toolkits.axes_grid1 import ImageGrid

from torchvision import transforms

import fire_network

import numpy as np



from PIL import Image

# Possible Scales for multiscale inference
scales = [2.0, 1.414, 1.0, 0.707, 0.5, 0.353, 0.25] 

device = 'cpu'

# Load net
state = torch.load('fire.pth', map_location='cpu')
state['net_params']['pretrained'] = None # no need for imagenet pretrained model
net = fire_network.init_network(**state['net_params']).to(device)
net.load_state_dict(state['state_dict'])

transform = transforms.Compose([
        transforms.Resize(1024),
        transforms.ToTensor(), 
        transforms.Normalize(**dict(zip(["mean", "std"], net.runtime['mean_std'])))
        ])


# which sf
sf_idx_ = [55, 14, 5, 4, 52, 57, 40, 9]

col = plt.get_cmap('tab10')

def generate_matching_superfeatures(im1, im2, scale_id=6, threshold=50):

    im1_tensor = transform(im1).unsqueeze(0)
    im2_tensor = transform(im2).unsqueeze(0)

    im1_cv = np.array(im1)[:, :, ::-1].copy() 
    im2_cv = np.array(im2)[:, :, ::-1].copy() 

    # extract features
    with torch.no_grad():
        output1 = net.get_superfeatures(im1_tensor.to(device), scales=[scale_id])
        feats1 = output1[0][0]
        attns1 = output1[1][0]
        strenghts1 = output1[2][0]

        output2 = net.get_superfeatures(im2_tensor.to(device), scales=[scale_id])
        feats2 = output2[0][0]
        attns2 = output2[1][0]
        strenghts2 = output2[2][0]

    print(feats1.shape, feats2.shape)
    print(attns1.shape, attns2.shape)
    print(strenghts1.shape, strenghts2.shape)
    
    # Store all binary SF att maps to show them all at once in the end
    all_att_bin1 = []
    all_att_bin2 = []
    for n, i in enumerate(sf_idx_):
        # all_atts[n].append(attn[j][scale_id][0,i,:,:].numpy())
        att_heat = np.array(attns1[0,i,:,:].numpy(), dtype=np.float32)
        att_heat = np.uint8(att_heat / np.max(att_heat[:]) * 255.0)
        att_heat_bin  = np.where(att_heat>threshold, 255, 0)
        print(att_heat_bin)
        all_att_bin1.append(att_heat_bin)

        att_heat = np.array(attns2[0,i,:,:].numpy(), dtype=np.float32)
        att_heat = np.uint8(att_heat / np.max(att_heat[:]) * 255.0)
        att_heat_bin  = np.where(att_heat>threshold, 255, 0)
        all_att_bin2.append(att_heat_bin)

    
    fin_img = []
    img1rsz = np.copy(im1_cv)
    print(im1.size)
    print(img1rsz.shape)
    for j, att in enumerate(all_att_bin1):
        att = cv2.resize(att, im1.size, interpolation=cv2.INTER_NEAREST)
        # att = cv2.resize(att, imgz[i].shape[:2][::-1], interpolation=cv2.INTER_CUBIC)
        # att = cv2.resize(att, imgz[i].shape[:2][::-1])
        # att = att.resize(shape)
        # att = resize(att, im1.size)
        mask2d = zip(*np.where(att==255))
        for m,n in mask2d:
            col_ = col.colors[j] if j < 7 else col.colors[j+1]
            if j == 0: col_ = col.colors[9]
            col_ = 255*np.array(colors.to_rgba(col_))[:3]
            img1rsz[m,n, :] = col_[::-1]   
    fin_img.append(img1rsz)
            
    img2rsz = np.copy(im2_cv)
    for j, att in enumerate(all_att_bin2):
        att = cv2.resize(att, im2.size, interpolation=cv2.INTER_NEAREST)
        # att = cv2.resize(att, imgz[i].shape[:2][::-1], interpolation=cv2.INTER_CUBIC)
        # # att = cv2.resize(att, imgz[i].shape[:2][::-1])
        # att = att.resize(im2.shape)
        # print('att:', att.shape)
        mask2d = zip(*np.where(att==255))
        for m,n in mask2d:
            col_ = col.colors[j] if j < 7 else col.colors[j+1]
            if j == 0: col_ = col.colors[9]
            col_ = 255*np.array(colors.to_rgba(col_))[:3]
            img2rsz[m,n, :] = col_[::-1]   
    fin_img.append(img2rsz)
    

    fig = plt.figure()
    grid = ImageGrid(fig, 111, nrows_ncols=(2, 1),  axes_pad=0.1)
    for ax, img in zip(grid, fin_img):
        ax.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
        ax.axis('scaled')
        ax.axis('off')
    plt.tight_layout()
    # fig.suptitle("Matching SFs", fontsize=16)

    # fig.canvas.draw()
    # # Now we can save it to a numpy array.
    # data = np.frombuffer(fig.canvas.tostring_rgb(), dtype=np.uint8)
    # data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))
    return fig


# GRADIO APP
title = "Visualizing Super-features"
description = "TBD"
article = "<p style='text-align: center'><a href='https://github.com/naver/fire' target='_blank'>Original Github Repo</a></p>"


# css = ".output-image, .input-image {height: 40rem !important; width: 100% !important;}"
# css = "@media screen and (max-width: 600px) { .output_image, .input_image {height:20rem !important; width: 100% !important;} }"
# css = ".output_image, .input_image {height: 600px !important}"
css = ".input_image {height: 600px !important} .output_image, {height: 1200px !important}"
# css = ".output-image, .input-image {height: 40rem !important; width: 100% !important;}"


iface = gr.Interface(
    fn=generate_matching_superfeatures,
    inputs=[
        gr.inputs.Image(shape=(1024, 1024), type="pil"),
        gr.inputs.Image(shape=(1024, 1024), type="pil"),
        gr.inputs.Slider(minimum=1, maximum=7, step=1, default=2, label="Scale"),
        gr.inputs.Slider(minimum=1, maximum=255, step=25, default=50, label="Binarizatio Threshold")],
    outputs="plot",
    # outputs=gr.outputs.Image(shape=(1024,2048), type="plot"),
    enable_queue=True,
    title=title,
    description=description,
    article=article,
    css=css,
    examples=[["chateau_1.png", "chateau_2.png", 6, 50]],
)
iface.launch()