eschmidbauer/DL-Art-School
1
0
Fork 0
forked from mrq/DL-Art-School
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ImageFolderDataset - corrupt lq images alongside each other

Browse Source
This commit is contained in:
James Betker 2021-01-03 16:36:38 -07:00
parent ce6524184c
commit 5e7ade0114
1 changed files with 22 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

41
codes/data/image_folder_dataset.py
View File

@ -12,6 +12,7 @@ from data import util
# Builds a dataset created from a simple folder containing a list of training/test/validation images.
from data.image_corruptor import ImageCorruptor
from data.image_label_parser import VsNetImageLabeler
from utils.util import opt_get
class ImageFolderDataset:
@ -25,8 +26,8 @@ class ImageFolderDataset:
self.corrupt_before_downsize = opt['corrupt_before_downsize'] if 'corrupt_before_downsize' in opt.keys() else False
self.fetch_alt_image = opt['fetch_alt_image'] # If specified, this dataset will attempt to find a second image
# from the same video source. Search for 'fetch_alt_image' for more info.
self.skip_lq = opt['skip_lq']
self.disable_flip = opt['disable_flip']
self.skip_lq = opt_get(opt, ['skip_lq'], False)
self.disable_flip = opt_get(opt, ['disable_flip'], False)
assert (self.target_hq_size // self.scale) % self.multiple == 0 # If we dont throw here, we get some really obscure errors.
if not isinstance(self.paths, list):
self.paths = [self.paths]
@ -119,16 +120,12 @@ class ImageFolderDataset:
hs = self.resize_hq([hq])
if not self.skip_lq:
ls = self.synthesize_lq(hs)
for_lq = [hs[0]]
# Convert to torch tensor
hq = torch.from_numpy(np.ascontiguousarray(np.transpose(hs[0], (2, 0, 1)))).float()
if not self.skip_lq:
lq = torch.from_numpy(np.ascontiguousarray(np.transpose(ls[0], (2, 0, 1)))).float()
out_dict = {'hq': hq, 'LQ_path': self.image_paths[item], 'HQ_path': self.image_paths[item]}
if not self.skip_lq:
out_dict['lq'] = lq
if self.fetch_alt_image:
# This works by assuming a specific filename structure as would produced by ffmpeg. ex:
@ -151,22 +148,28 @@ class ImageFolderDataset:
# the file rather than searching the path list. Let the exception handler below do its work.
next_img = self.image_paths[item].replace(str(imnumber), str(imnumber+1))
alt_hq = util.read_img(None, next_img, rgb=True)
alt_hq = self.resize_hq([alt_hq])
alt_hq = torch.from_numpy(np.ascontiguousarray(np.transpose(alt_hq[0], (2, 0, 1)))).float()
alt_hs = self.resize_hq([alt_hq])
alt_hq = torch.from_numpy(np.ascontiguousarray(np.transpose(alt_hs[0], (2, 0, 1)))).float()
if not self.skip_lq:
alt_lq = self.synthesize_lq(alt_hq)
alt_lq = torch.from_numpy(np.ascontiguousarray(np.transpose(alt_lq[0], (2, 0, 1)))).float()
for_lq.append(alt_hs[0])
except:
alt_hq = hq
if not self.skip_lq:
alt_lq = lq
for_lq.append(hs[0])
else:
alt_hq = hq
if not self.skip_lq:
alt_lq = lq
for_lq.append(hs[0])
out_dict['alt_hq'] = alt_hq
if not self.skip_lq:
out_dict['alt_lq'] = alt_lq
if not self.skip_lq:
lqs = self.synthesize_lq(for_lq)
ls = lqs[0]
out_dict['lq'] = torch.from_numpy(np.ascontiguousarray(np.transpose(ls, (2, 0, 1)))).float()
if len(lqs) > 1:
alt_lq = lqs[1]
out_dict['alt_lq'] = torch.from_numpy(np.ascontiguousarray(np.transpose(alt_lq, (2, 0, 1)))).float()
if self.labeler:
base_file = self.image_paths[item].replace(self.paths[0], "")
@ -190,7 +193,7 @@ if __name__ == '__main__':
'fixed_corruptions': ['jpeg-broad', 'gaussian_blur'],
'random_corruptions': ['noise-5', 'none'],
'num_corrupts_per_image': 1,
'corrupt_before_downsize': True,
'corrupt_before_downsize': False,
'fetch_alt_image': True,
#'labeler': {
# 'type': 'patch_labels',
@ -203,11 +206,11 @@ if __name__ == '__main__':
os.makedirs("debug", exist_ok=True)
for i in range(0, len(ds)):
o = ds[random.randint(0, len(ds)-1)]
hq = o['hq']
hq = o['lq']
#masked = (o['labels_mask'] * .5 + .5) * hq
import torchvision
torchvision.utils.save_image(hq.unsqueeze(0), "debug/%i_hq.png" % (i,))
torchvision.utils.save_image(o['alt_hq'].unsqueeze(0), "debug/%i_hq_alt.png" % (i,))
torchvision.utils.save_image(hq.unsqueeze(0), "debug/%i_lq.png" % (i,))
torchvision.utils.save_image(o['alt_lq'].unsqueeze(0), "debug/%i_lq_alt.png" % (i,))
#if len(o['labels'].unique()) > 1:
# randlbl = np.random.choice(o['labels'].unique()[1:])
# moremask = hq * ((1*(o['labels'] == randlbl))*.5+.5)