-
Notifications
You must be signed in to change notification settings - Fork 16
/
config.py
44 lines (33 loc) · 1.64 KB
/
config.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
"""
some configuration set here
"""
import numpy as np
dataset_name = 'inria_person' ##can be 'hazy_person or inria_person'
category_index = {0: {"name": "Background"},
1: {"name": "Person"}}
# img_size = (224,224) ##deafult img size---(h, w)
img_size = (448,448) ##deafult img size---(h, w)
## for hazy person
# priori_bboxes = np.array([[158, 65], [65, 79]], dtype=np.float32) ##this priori bboxes are produced by k-means,encoded by [height, width]
# priori_bboxes = np.array([[58, 146], [155, 63], [53, 24]], dtype=np.float32)
# priori_bboxes = np.array([[87, 59], [192, 87], [107, 243], [310, 115], [255, 175], [323, 270]], dtype=np.float32)*(448/416)
# priori_bboxes = np.array([[162, 72], [108, 237], [266, 129], [312, 244]], dtype=np.float32)*(448/416)
## for inria person
priori_bboxes = np.array([[134, 47], [72, 22]], dtype=np.float32)
# priori_bboxes = np.array([[83, 31], [153, 48], [78,113], [76,235], [118, 173], [166, 269]], dtype=np.float32)*(448/416)
# priori_bboxes = np.array([[85, 35], [137, 54], [192,91], [258,159]], dtype=np.float32)*(448/416)
## for union_person
# priori_bboxes = np.array([[73, 32], [139, 87]], dtype=np.float32)
grid_cell_size = (14, 14) ##grid cell size, means divide img into 7x7 grid cells
top_k = 2 ##choose top-k as positive sample
surounding_size = 4 ##surounding_size = 4
# img_size = (448,448)
# priori_bboxes = np.array([[158, 65], [65, 79]], dtype=np.float32)*2
# grid_cell_size = (14, 14)
# top_k = 2
# surounding_size = 5
# img_size = (320,320)
# priori_bboxes = np.array([[158, 65], [65, 79]], dtype=np.float32)*1.4285
# grid_cell_size = (10, 10)
# top_k = 1
# surounding_size = 5