-
Notifications
You must be signed in to change notification settings - Fork 6
/
demo_scoring.py
157 lines (130 loc) · 7.34 KB
/
demo_scoring.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
"""
ArTIST evaluation: Inferring the likelihood of current observations (detections)
author: Fatemeh Saleh
"""
import cv2
from utils.utils_ar import infer_log_likelihood, iou
import numpy as np
import torch
from utils.clustering import clustering, load_clusters
from models.ar import motion_ar
from models.ae import motion_ae
torch.backends.cudnn.enabled = False
def test(model, centroid_x, centroid_y, centroid_w, centroid_h, cnt, num_cluster):
"""[summary]
Args:
model (nn.Module): ArTIST model
centroid_x (list): The centroids of x coordinate of the bounding boxes in the training set
centroid_y (list): The centroids of y coordinate of the bounding boxes in the training set
centroid_w (list): The centroids of width of the bounding boxes in the training set
centroid_h ((list): The centroids of height of the bounding boxes in the training set
cnt (int): The test sequence index
num_cluster (int): number of clusters
"""
test_set = np.load('data/demo_test_subset.npy', allow_pickle=True)
rand_len = int(test_set[cnt]['seq_len'])
# we consider the sequence to be observed up until mask_index
mask_index = int(rand_len * 0.75)
data = test_set[cnt]['data'][:, :mask_index, :]
BBOX = []
image_wh = test_set[cnt]['wh'][0]
width = image_wh[0]
height = image_wh[1]
# social information (see section 3.3 of https://arxiv.org/pdf/2012.02337v1.pdf)
social = test_set[cnt]['social']
valid_box = test_set[cnt]['data'][:, mask_index, :]
# creating some imaginary detections at current time step
instance_options = [[820/width, 830/height, 80/width, 170/height], [400/width, 310/height, 110/width, 330/height], [1000/width, 300/height, 170/width, 300/height]]
instance_options.append([valid_box[0, 0], valid_box[0, 1], valid_box[0, 2], valid_box[0, 3]])
scores, ious = np.zeros(len(instance_options)), np.zeros(len(instance_options))
gap = 1
gaussian_kernel = np.load("data/kernel.npy")
gaussian_kernel = torch.autograd.Variable(torch.from_numpy(gaussian_kernel).float()).cuda()
gaussian_kernel = gaussian_kernel.unsqueeze(0).unsqueeze(0)
self_data = data.cuda()
self_data = self_data.cuda().float()
# computing the motion velocity
self_delta_tmp = self_data[:, 1:, :] - self_data[:, 0:-1, :]
self_delta = torch.zeros(self_delta_tmp.shape[0], self_delta_tmp.shape[1] + 1,
self_delta_tmp.shape[2]).cuda()
self_delta[:, 1:, :] = self_delta_tmp
# computing the distribution over the next plausible bounding box
dist_x, dist_y, dist_w, dist_h, sampled_boxes, sampled_deltas, sampled_detection = model.inference(
self_data,
social[:, :mask_index, :],
gap,
centroid_x,
centroid_y,
centroid_w,
centroid_h)
# making it a probability distribution
dist_x = torch.nn.Softmax(dim=-1)(dist_x)
dist_y = torch.nn.Softmax(dim=-1)(dist_y)
dist_w = torch.nn.Softmax(dim=-1)(dist_w)
dist_h = torch.nn.Softmax(dim=-1)(dist_h)
# smoothing the distributions using a gaussian kernel
y_g1d_x = torch.nn.functional.conv1d(dist_x, gaussian_kernel.repeat(dist_x.shape[1], dist_x.shape[1], 1),
padding=24)
y_g1d_y = torch.nn.functional.conv1d(dist_y, gaussian_kernel.repeat(dist_y.shape[1], dist_y.shape[1], 1),
padding=24)
y_g1d_w = torch.nn.functional.conv1d(dist_w, gaussian_kernel.repeat(dist_w.shape[1], dist_w.shape[1], 1),
padding=24)
y_g1d_h = torch.nn.functional.conv1d(dist_h, gaussian_kernel.repeat(dist_h.shape[1], dist_h.shape[1], 1),
padding=24)
extended_track = torch.zeros(1, self_delta.shape[1] + len(sampled_boxes) + 1, 4).cuda()
extended_track[0, :self_delta.shape[1], :] = self_delta[0, :, :]
extended_track[0, self_delta.shape[1]:-1, :] = sampled_deltas[0, :-1, :]
observation_last = [self_data[0, -1, 0].item() * width, self_data[0, -1, 1].item() * height,
(self_data[0, -1, 0].item() * width) + (self_data[0, -1, 2].item() * width),
(self_data[0, -1, 1].item() * height) + (self_data[0, -1, 3].item() * height)]
# loop over the detections...
for opt_idx, option in enumerate(instance_options):
opt = [(option[0]), (option[1]), (option[2]), (option[3])]
option_unnorm = [
option[0] * width,
option[1] * height,
(option[2] + option[0]) * width,
(option[3] + option[1]) * height
]
iou_validate = iou(option_unnorm, observation_last)
ious[opt_idx] = iou_validate
last_delta = torch.from_numpy(np.array(opt) - self_data[0, -1].cpu().detach().numpy()).cuda()
extended_track[0, -1, :] = last_delta
# inferring the likelihood of each detection (considered as the last observation of the sequence)
likelihoods_smooth = infer_log_likelihood(y_g1d_x, y_g1d_y, y_g1d_w, y_g1d_h,
extended_track[:, 1:, 0:1],
extended_track[:, 1:, 1:2],
extended_track[:, 1:, 2:3],
extended_track[:, 1:, 3:4],
centroid_x, centroid_y, centroid_w, centroid_h, num_cluster)
all_scores = np.array(likelihoods_smooth[-1])
likelihoods_smooth = np.sum(all_scores)
score = likelihoods_smooth
scores[opt_idx] = score
# visualization
I1 = np.ones((int(image_wh[1].item()), int(image_wh[0].item()), 3))
I1 = I1 * 255
for i in range(0, mask_index):
I1 = cv2.rectangle(I1, (
int(data[0, i, 0].item() * image_wh[0]), int(data[0, i, 1].item() * image_wh[1])),
(int(data[0, i, 0].item() * image_wh[0] + data[0, i, 2].item() * image_wh[0]),
int(data[0, i, 1].item() * image_wh[1] + data[0, i, 3].item() * image_wh[1])),
(255, 0, 0), 1) #### GT
for i in range(len(instance_options)):
I1 = cv2.rectangle(I1, (
int(instance_options[i][0]*width), int(instance_options[i][1]*height),
int(instance_options[i][2]*width), int(instance_options[i][3]*height)), (0, 0, 0), 2)
cv2.putText(I1, 'log p: ' + str(round(scores[i], 2)), (int(instance_options[i][0]*width), int(instance_options[i][1]*height)-40), cv2.FONT_HERSHEY_PLAIN, 1.5, (0, 0, 0),
1, lineType=cv2.LINE_AA)
cv2.putText(I1, 'IoU w/ last bbox: ' + str(round(ious[i], 2)), (int(instance_options[i][0]*width), int(instance_options[i][1]*height)-15), cv2.FONT_HERSHEY_PLAIN, 1.5, (0, 0, 0),
1 , lineType=cv2.LINE_AA)
cv2.imwrite('temp/ar/log_p/likelihood.jpg', I1)
if __name__ == '__main__':
model_ae = motion_ae(256).cuda()
model_ae.load_state_dict(torch.load('checkpoint/ae/ae_8.pth'))
model_ae.eval()
model_ar = motion_ar(512, 1024).cuda()
model_ar.load_state_dict(torch.load('checkpoint/ar/ar_110.pth'))
model_ar.eval()
centroid_x, centroid_y, centroid_w, centroid_h = load_clusters()
test(model_ar, centroid_x, centroid_y, centroid_w, centroid_h, 100, 1024)