def __init__(self, opt):
super(LRHRfromtxt, self).__init__()
self.opt = opt
self.train = (opt['phase'] == 'train')
self.split = 'train' if self.train else 'test'
self.scale = self.opt['scale']
self.paths_HR, self.paths_LR = None, None
self.qualityscore = None
self.list_LR = []
self.list_HR = []
# change the length of train dataset (influence the number of iterations in each epoch)
self.repeat = 2
# read image list from image/binary files
self.paths_LR, self.paths_HR, self.qualityscore = common.get_image_paths_from_txt(self.opt['txtpath'],1)
for i in range(len(self.paths_LR)):
lr_path = self.paths_LR[i]
hr_path = self.paths_HR[i]
lr = common.read_img_fromtxt(lr_path, self.opt['data_type'])
hr = common.read_img_fromtxt(hr_path, self.opt['data_type'])
self.list_LR.append(lr)
self.list_HR.append(hr)
assert self.paths_HR, '[Error] HR paths are empty.'
if self.paths_LR and self.paths_HR:
assert len(self.paths_LR) == len(self.paths_HR), \
'[Error] HR: [%d] and LR: [%d] have different number of images.'%(
len(self.paths_LR), len(self.paths_HR))
def _load_file(self, idx):
lr_path1 = self.paths_LR1[idx]
lr_path2 = self.paths_LR2[idx]
lr_path3 = self.paths_LR3[idx]
lr1 = common.read_img_fromtxt(lr_path1, self.opt['data_type'])
lr2 = common.read_img_fromtxt(lr_path2, self.opt['data_type'])
lr3 = common.read_img_fromtxt(lr_path3, self.opt['data_type'])
return lr1, lr2, lr3, lr_path1, lr_path2, lr_path3
def _load_file(self, idx):
idx = self._get_index(idx)
lr_path1 = self.paths_LR1[idx]
lr_path2 = self.paths_LR2[idx]
lr_path3 = self.paths_LR3[idx]
hr_path = self.paths_HR[idx]
lr1 = common.read_img_fromtxt(lr_path1, self.opt['data_type'])
lr2 = common.read_img_fromtxt(lr_path2, self.opt['data_type'])
lr3 = common.read_img_fromtxt(lr_path3, self.opt['data_type'])
hr = common.read_img_fromtxt(hr_path, self.opt['data_type'])
return lr1,lr2, lr3, hr, lr_path1, lr_path2, lr_path3, hr_path