-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathvisualize.py
192 lines (151 loc) · 6.23 KB
/
visualize.py
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
#!/usr/bin/env python3
"""
Amit Kohli, Julien Martel, and Anastasios Angelopoulos
August 10, 2020
This script provides a visualization of the ebv-eye data.
"""
import argparse
import struct
import glob
import os
import matplotlib.pyplot as plt
from PIL import Image
from collections import namedtuple
parser = argparse.ArgumentParser(description='Arguments for using the eye visualizer')
parser.add_argument('--subject', type=int, default=22, help='which subject to evaluate')
parser.add_argument('--eye', default='left', choices=['left', 'right'],
help='Which eye to visualize, left or right')
parser.add_argument('--data_dir', default=os.path.join(os.getcwd(), 'eye_data'),
help='absolute path to eye_data/, by default assumes same parent dir as this script')
parser.add_argument('--buffer', type=int, default=1000, help='How many events to store before displaying.')
opt = parser.parse_args()
'Types of data'
Event = namedtuple('Event', 'polarity row col timestamp')
Frame = namedtuple('Frame', 'row col img timestamp')
'Color scheme for event polarity'
color = ['r', 'g']
def glob_imgs(path):
imgs = []
for ext in ['*.png', '*.jpg', '*.JPEG', '*.JPG']:
imgs.extend(glob.glob(os.path.join(path,'**', ext), recursive=True))
return imgs
'Reads an event file'
def read_aerdat(filepath):
with open(filepath, mode='rb') as file:
file_content = file.read()
''' Packet format'''
packet_format = 'BHHI' # pol = uchar, (x,y) = ushort, t = uint32
packet_size = struct.calcsize('='+packet_format) # 16 + 16 + 8 + 32 bits => 2 + 2 + 1 + 4 bytes => 9 bytes
num_events = len(file_content)//packet_size
extra_bits = len(file_content)%packet_size
'''Remove Extra Bits'''
if extra_bits:
file_content = file_content[0:-extra_bits]
''' Unpacking'''
event_list = list(struct.unpack('=' + packet_format * num_events, file_content))
event_list.reverse()
return event_list
'Parses the filename of the frames'
def get_path_info(path):
path = path.split('/')[-1]
filename = path.split('.')[0]
path_parts = filename.split('_')
index = int(path_parts[0])
stimulus_type = path_parts[3]
timestamp = int(path_parts[4])
return {'index': index, 'row': int(path_parts[1]), 'col': int(path_parts[2]), 'stimulus_type': stimulus_type,
'timestamp': timestamp}
'Manages both events and frames as a general data object'
class EyeDataset:
'Initialize by creating a time ordered stack of frames and events'
def __init__(self, data_dir, user):
self.data_dir = data_dir
self.user = user
self.frame_stack = []
self.event_stack = []
def __len__(self):
return len(self.frame_stack) + len(self.event_stack)
def __getitem__(self, index):
'Determine if event or frame is next in time by peeking into both stacks'
frame_timestamp = self.frame_stack[-1].timestamp
event_timestamp = self.event_stack[-4]
'Returns selected data type'
if event_timestamp < frame_timestamp:
polarity = self.event_stack.pop()
row = self.event_stack.pop()
col = self.event_stack.pop()
timestamp = self.event_stack.pop()
event = Event(polarity, row, col, timestamp)
return event
else:
frame = self.frame_stack.pop()
img = Image.open(frame.img).convert("L")
frame = frame._replace(img=img)
return frame
'Loads in data from the data_dir as filenames'
def collect_data(self, eye=0):
print('Loading Frames....')
self.frame_stack = self.load_frame_data(eye)
print('There are ' + str(len(self.frame_stack)) + ' frames \n')
print('Loading Events....')
self.event_stack = self.load_event_data(eye)
print('There are ' + str(len(self.event_stack)) + ' events \n')
def load_frame_data(self, eye):
filepath_list = []
user_name = "user" + str(self.user)
img_dir = os.path.join(self.data_dir, user_name, str(eye), 'frames')
img_filepaths = list(glob_imgs(img_dir))
img_filepaths.sort(key=lambda name: get_path_info(name)['index'])
img_filepaths.reverse()
for fpath in img_filepaths:
path_info = get_path_info(fpath)
frame = Frame(path_info['row'], path_info['col'], fpath, path_info['timestamp'])
filepath_list.append(frame)
return filepath_list
def load_event_data(self, eye):
user_name = "user" + str(self.user)
event_file = os.path.join(self.data_dir, user_name, str(eye), 'events.aerdat')
filepath_list = read_aerdat(event_file)
return filepath_list
'Displays the data as fast as GUI can render'
def display_data(eye_dataset):
col_buffer = []
row_buffer = []
polarity_buffer = []
s = plt.plot([],[])[0]
init_img_axis = False
for i, data in enumerate(eye_dataset):
if type(data) is Frame:
if not init_img_axis:
img_axis = plt.imshow(data.img)
init = True
else:
img_axis.set_data(data.img)
plt.draw()
plt.pause(0.0001)
else:
col_buffer += [data.col]
row_buffer += [data.row]
polarity_buffer += [color[data.polarity]]
if not len(col_buffer) % opt.buffer:
s.remove()
s = plt.scatter(col_buffer, row_buffer, color=polarity_buffer, s=1)
col_buffer.clear()
row_buffer.clear()
polarity_buffer.clear()
plt.pause(0.0001)
plt.show()
def main():
eye_dataset = EyeDataset(opt.data_dir, opt.subject)
if opt.eye == 'left':
print('Showing the left eye of subject ' + str(opt.subject) + '\n')
print('Loading Data from ' + opt.data_dir + '..... \n')
eye_dataset.collect_data(0)
else:
print('Showing the right eye of subject ' + str(opt.subject)+ '\n')
print('Loading Data from ' + opt.data_dir + '..... \n')
eye_dataset.collect_data(1)
print('Displaying Data using a ' + str(opt.buffer)+ ' event buffer...')
display_data(eye_dataset)
if __name__ == '__main__':
main()