-
Notifications
You must be signed in to change notification settings - Fork 10
/
Copy pathPlotExtractorIoU.py
180 lines (135 loc) · 5.68 KB
/
PlotExtractorIoU.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
import os
import numpy as np
import matplotlib.pyplot as plt
import ast
import cv2
import argparse
from joblib import load
from matplotlib import rcParams
rcParams['font.family'] = ['serif']
rcParams['font.serif'] = ['Times New Roman']
rcParams['font.size'] = 16
parser = argparse.ArgumentParser()
parser.add_argument("-v", "--video", default=11,
help="Index of the video")
parser.add_argument("-m", "--metric", default='euclidean',
help="metric for the tracking backend")
args = parser.parse_args()
video = f"source_material/streetVideos/seattle{args.video}.mp4"
videoDir = f"results/plots_tables/{args.metric}/video{args.video}"
csvDirSamples = f"results/plots_tables/{args.metric}/samples.txt"
csvDirSpeeds = f"results/plots_tables/{args.metric}/speeds.txt"
csvDirLineSpeeds = f"results/plots_tables/{args.metric}/lineSpeeds.txt"
# Create containing folder for plots
if not os.path.exists(videoDir):
os.mkdir(videoDir)
# Create file for numeric results
f_samples = open(csvDirSamples, 'a+')
f_speeds = open(csvDirSpeeds, 'a+')
f_lineSpeeds = open(csvDirLineSpeeds, 'a+')
# Load video to get FPS
cap = cv2.VideoCapture(video)
if (cap.isOpened()== False):
raise Exception("Error opening video stream or file")
fps = int(cap.get(cv2.CAP_PROP_FPS))
# Load all result data from video execution
bbDists = load(f"results/bbox_distances/{args.metric}/bbox_distances{args.video}")
bbDists = ast.literal_eval(bbDists)
data = load(f"results/velocities/{args.metric}/velocities{args.video}")
spotVelocities = load(f"results/spot_velocities/{args.metric}/spot_velocities{args.video}")
all_lengths = list(map(lambda arr: len(arr), data))
print("saving sampling data")
registered_objects = len(all_lengths)
mean_samples = np.mean(all_lengths)
median_samples = np.median(all_lengths)
std_samples = np.std(all_lengths)
f_samples.write(f"{registered_objects},{mean_samples},{median_samples},{std_samples}\n")
print("saving sampling plot")
_ = plt.hist(all_lengths, bins=50, color='g')
plt.yscale('log')
_ = plt.xlabel("n_samples")
_ = plt.ylabel("Frequency")
plt.savefig(f"{videoDir}/samplesPerObject{args.video}.jpg")
print("saving speed data")
all_samples_concat = np.hstack(list(data))
len_all_samples = len(all_samples_concat)
mean_speeds = np.mean(all_samples_concat)
median_speeds = np.median(all_samples_concat)
std_speeds = np.std(all_samples_concat)
portion_70 = len(all_samples_concat[all_samples_concat > 70])/len(all_samples_concat)
portion_100 = len(all_samples_concat[all_samples_concat > 100])/len(all_samples_concat)
f_speeds.write(f"{len_all_samples},{mean_speeds},{median_speeds},{std_speeds},{portion_70},{portion_100}\n")
print("saving speed plots")
plt.clf()
_ = plt.hist(all_samples_concat, bins=100, density=False)
_ = plt.axvline(x=np.mean(all_samples_concat), color='red')
plt.yscale("log")
plt.xlabel("Speed (km/h)")
plt.ylabel("Frequency")
plt.savefig(f"{videoDir}/SpeedDistribution{args.video}.jpg")
sorted_samples = sorted(data, reverse=True, key=lambda arr: len(arr))[:10]
sorted_lens = list(map(lambda arr: len(arr), sorted_samples))
median_lens, std_lens = np.median(sorted_lens), np.std(sorted_lens)
inds = [list(map(lambda t: t*25/fps,range(len(sorted_sample)))) for sorted_sample in sorted_samples]
sorted_bbs = sorted(bbDists, reverse=True, key=lambda arr: len(arr))[:10]
sorted_lens_bb = list(map(lambda arr: len(arr), sorted_bbs))
median_lens_bb, std_lens_bb = np.median(sorted_lens_bb), np.std(sorted_lens_bb)
inds_bbs = [range(len(sorted_bb)) for sorted_bb in sorted_bbs]
print("saving speed time series")
plt.clf()
for i in range(len(inds)):
_ = plt.plot(inds[i], sorted_samples[i], )
_ = plt.ylim(0,100)
_ = plt.xlabel("Time (s)")
_ = plt.ylabel("Speed (km/h)")
plt.savefig(f"{videoDir}/SpeedTimeSeries.jpg")
print("saving separated speed time series")
plt.clf()
f, ax = plt.subplots(len(inds), 1, sharex=False, sharey=False)
for i in range(len(inds)):
ax[i].plot(inds[i], sorted_samples[i])
f.text(0.5, 0.04, 'Time (s)', ha='center')
f.text(0.01, 0.5, 'Speed (km/h)', va='center', rotation='vertical')
plt.subplots_adjust(hspace=0.5)
f.set_size_inches(6.5, 9.5)
plt.savefig(f"{videoDir}/multiPlotTimeSeries{args.video}.jpg")
print("saving bounding box time series")
plt.clf()
for i in range(len(inds)):
_ = plt.plot(inds_bbs[i], sorted_bbs[i])
_ = plt.xlabel("Frames")
_ = plt.ylabel("Diagonal size")
plt.savefig(f"{videoDir}/bbDistTimeSeries.jpg")
print("saving bounding box separated time series")
plt.clf()
f, ax = plt.subplots(len(inds_bbs), 1, sharex=False, sharey=False)
for i in range(len(inds_bbs)):
ax[i].plot(inds_bbs[i], sorted_bbs[i])
f.text(0.5, 0.04, 'Time (s)', ha='center')
f.text(0.04, 0.5, 'Speed (km/h)', va='center', rotation='vertical')
plt.subplots_adjust(hspace=0.5)
f.set_size_inches(6.5, 9.5)
plt.savefig(f"{videoDir}/multiPlotBBsTimeSeries.jpg")
print("saving speed boxplots")
plt.clf()
fig, ax = plt.subplots()
_ = ax.boxplot(np.array(sorted_samples, dtype=object))
_ = ax.set_xlabel("object ID")
_ = ax.set_ylabel("Speed (km/h)")
plt.savefig(f"{videoDir}/speedBoxPlots{args.video}.jpg")
print("saving line speeds data")
registered_line_speeds = len(spotVelocities)
mean_line_speed = np.mean(spotVelocities)
median_line_speed = np.median(spotVelocities)
std_line_speed = np.std(spotVelocities)
f_lineSpeeds.write(f"{registered_line_speeds},{mean_line_speed},{median_line_speed},{std_line_speed}\n")
print("saving line speed distribution")
plt.clf()
_ = plt.hist(spotVelocities, bins=10, color="#FFA500")
_ = plt.xlabel("Speed (km/h)")
_ = plt.ylabel("Frequency")
plt.savefig(f"{videoDir}/SpeedLineDistribution{args.video}.jpg")
f_samples.close()
f_speeds.close()
f_lineSpeeds.close()
print(f"All done for video {args.video} with {args.metric}!")