Estimating depth from images

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
#!/usr/bin/env python
# Example script to load the data, visualize the images, and capture clicks to identify the book
# Run the script as: ./process_images.py images.npz

import sys
import cv2
import numpy as np
from extract_coordinates import extract_coordinates

# -------------------------------------------------------------------------------
# get input path
images_file = sys.argv[1]

# load data
print("Loading {}".format(images_file))
data = np.load(images_file)

gray=data['gray']              # gray image captured by the camera
depth=data['depth']            # aligned depth image (transformed to match the gray image)
width=data['width']            # images width
height=data['height']          # images height
K=np.reshape(data['K'], (3,3)) # camera intrinsic parameters

# -------------------------------------------------------------------------------
# display the depth image
# but first normalize it so that all parts are visible and not too dark...
scaled_depth = 255 * depth / depth.max() # normalize data to 0 to 255
depth_as_int = scaled_depth.astype(np.uint8)

cv2.imshow("Depth", depth_as_int)
cv2.moveWindow("Depth", 20,20);

# display the gray image
cv2.imshow("Gray", gray)
cv2.moveWindow("Gray", 20 + width, 20)

print("You should now see the depth and gray image. Press any key to continue...")
cv2.waitKey(0)

# close all windows
cv2.destroyAllWindows() 

# -------------------------------------------------------------------------------
# get coordinates of book in the gray image
image_coordinates = []
image = gray 

cv2.namedWindow('select')
cv2.setMouseCallback('select', extract_coordinates, param=(image_coordinates, image))
cv2.imshow("select", gray)

print("Identify the object of interest by clicking the top-left corner of the front cover of the book with the left mouse button, holding the press until the mouse is in the bottom-right corner of the book, and then releasing the left button...")

while len(image_coordinates) < 2:
    cv2.waitKey(1) # wait until the region of interest is selected on the image...

cv2.destroyAllWindows()

print("Top-left: {}".format(image_coordinates[0]))
print("Bottom-right: {}".format(image_coordinates[1]))

# -------------------------------------------------------------------------------
# Implement your code donw below...

## pV-2
object_height = float(sys.argv[2])

x1, y1 = image_coordinates[0]
x2, y2 = image_coordinates[1]

Y1z = (y1-K[1][2])/K[1][1]
Y2z = (y2-K[1][2])/K[1][1]

Z = object_height/abs(Y1z-Y2z)
print("Estimated depth: {}m".format(Z))


## pV-3
cnt = 0
tot = 0
for r in range(y1, y2+1):
    for c in range(x1, x2+1):
        if depth[r][c]== 0.:
            continue
        cnt += 1
        tot += depth[r][c]
average_depth = tot/cnt
print("Estimated depth from depth image: {}m".format(average_depth)) # where average_depth is computed from the depth image
print("Difference between estimates: {}".format(average_depth - Z))