search_box 怎么理解 生成box 训练对应坐标关系 gen_json.py par_crop.py

[wavelet2008]

yt_bb

11/-0oXPaBxxxE": 000034.01.x.jpg
"000034": [
380.64,
16.2,
480.0,
184.79998799999998
]
search_image= cv2.imread("E:\pyth3\pysot\tools\xx\000034.01.x.jpg")
search_box = _get_bbox(search_image, [380,16.2,480,184])
<search_box >: (231.0959791303944, 214.88905298080178, 278.90402086960563, 295.1109470191982)
怎么理解 生成box 训练对应坐标关系
def _get_bbox( image, shape):
imh, imw = image.shape[:2]
if len(shape) == 4:
w, h = shape[2]-shape[0], shape[3]-shape[1]
else:
w, h = shape
context_amount = 0.5
exemplar_size = cfg.TRAIN.EXEMPLAR_SIZE
wc_z = w + context_amount * (w+h)
hc_z = h + context_amount * (w+h)
s_z = np.sqrt(wc_z * hc_z)
scale_z = exemplar_size / s_z
w = wscale_z
h = hscale_z
cx, cy = imw//2, imh//2
bbox = center2corner(Center(cx, cy, w, h))
return bbox

[wavelet2008]

我感觉知道了