def upscale_image(cfg, src, scale_model, alpha_model=None):
dst, alpha = split_alpha(src, scale_model)
for i in range(int(np.ceil(np.log2(cfg.scale_ratio)))):
six.print_('2.0x upscaling...', end=' ', flush=True)
model = scale_model if i == 0 or alpha_model is None else alpha_model
if model.inner_scale == 1:
dst = iproc.nn_scaling(dst, 2) # Nearest neighbor 2x scaling
alpha = iproc.nn_scaling(alpha, 2) # Nearest neighbor 2x scaling
if cfg.tta:
dst = reconstruct.image_tta(dst, model, cfg.tta_level,
cfg.block_size, cfg.batch_size)
else:
dst = reconstruct.image(dst, model, cfg.block_size, cfg.batch_size)
if alpha_model is None:
alpha = reconstruct.image(alpha, scale_model, cfg.block_size,
cfg.batch_size)
else:
alpha = reconstruct.image(alpha, alpha_model, cfg.block_size,
cfg.batch_size)
six.print_('OK')
dst_w = int(np.round(src.size[0] * cfg.scale_ratio))
dst_h = int(np.round(src.size[1] * cfg.scale_ratio))
if dst_w != dst.size[0] or dst_h != dst.size[1]:
six.print_('Resizing...', end=' ', flush=True)
dst = dst.resize((dst_w, dst_h), Image.LANCZOS)
six.print_('OK')
if alpha is not None:
if alpha.size[0] != dst_w or alpha.size[1] != dst_h:
alpha = alpha.resize((dst_w, dst_h), Image.LANCZOS)
dst.putalpha(alpha)
return dst
def denoise_image(src, model, cfg):
six.print_('Level %d denoising...' % cfg.noise_level, end=' ', flush=True)
if cfg.tta:
dst = reconstruct.image_tta(src, model, cfg.tta_level, cfg.block_size,
cfg.batch_size)
else:
dst = reconstruct.image(src, model, cfg.block_size, cfg.batch_size)
six.print_('OK')
return dst
def upscale_image(cfg, src, model):
dst = src.copy()
six.print_('2.0x scaling...', end=' ', flush=True)
if model.inner_scale == 1:
dst = iproc.nn_scaling(dst, 2) # Nearest neighbor 2x scaling
if cfg.tta:
dst = reconstruct.image_tta(dst, model, cfg.tta_level, cfg.block_size,
cfg.batch_size)
else:
dst = reconstruct.image(dst, model, cfg.block_size, cfg.batch_size)
six.print_('OK')
return dst
def denoise_image(cfg, src, model):
dst = src.copy()
six.print_('Level {} denoising...'.format(cfg.noise_level),
end=' ',
flush=True)
if cfg.tta:
dst = reconstruct.image_tta(dst, model, cfg.tta_level, cfg.block_size,
cfg.batch_size)
else:
dst = reconstruct.image(dst, model, cfg.block_size, cfg.batch_size)
if model.inner_scale != 1:
dst = dst.resize((src.size[0], src.size[1]), Image.LANCZOS)
six.print_('OK')
return dst
def denoise_image(src, model, cfg):
dst, alpha = split_alpha(src, model.offset)
six.print_('Level %d denoising...' % cfg.noise_level, end=' ', flush=True)
if cfg.tta:
dst = reconstruct.image_tta(dst, model, cfg.tta_level, cfg.block_size,
cfg.batch_size)
else:
dst = reconstruct.image(dst, model, cfg.block_size, cfg.batch_size)
if model.inner_scale != 1:
dst = dst.resize((src.size[0], src.size[1]), Image.LANCZOS)
six.print_('OK')
if alpha is not None:
dst.putalpha(alpha)
return dst
def upscale_image(src, model, cfg):
dst = src
if model.inner_scale == 1:
dst = src.resize((src.size[0] * 2, src.size[1] * 2), Image.NEAREST)
log_scale = np.log2(cfg.scale_factor)
for _ in range(int(np.ceil(log_scale))):
six.print_('2.0x upscaling...', end=' ', flush=True)
if cfg.tta:
dst = reconstruct.image_tta(dst, model, cfg.tta_level,
cfg.block_size, cfg.batch_size)
else:
dst = reconstruct.image(dst, model, cfg.block_size, cfg.batch_size)
six.print_('OK')
if np.round(log_scale % 1.0, 6) != 0:
six.print_('Resizing...', end=' ', flush=True)
dst_w = int(np.round(src.size[0] * cfg.scale_factor))
dst_h = int(np.round(src.size[1] * cfg.scale_factor))
dst = dst.resize((dst_w, dst_h), Image.ANTIALIAS)
six.print_('OK')
return dst
