def process(img):
img = bytes_to_file(img)
try:
img_A, img_B = read_split_image(img)
if augment:
# augment the image by:
# 1) enlarge the image
# 2) random crop the image back to its original size
# NOTE: image A and B needs to be in sync as how much
# to be shifted
w, h = img_A.shape
multiplier = random.uniform(1.00, 1.05)
# add an eps to prevent cropping issue
nw = int(multiplier * w) + 1
nh = int(multiplier * h) + 1
shift_x = int(np.ceil(np.random.uniform(0.01, nw - w)))
shift_y = int(np.ceil(np.random.uniform(0.01, nh - h)))
img_A = shift_and_resize_image(img_A, shift_x, shift_y, nw, nh)
img_B = shift_and_resize_image(img_B, shift_x, shift_y, nw, nh)
img_A = normalize_image(img_A)
img_B = normalize_image(img_B)
merged = np.stack([img_A, img_B], axis=2)
return merged
finally:
img.close()
super(NeverEndingLoopingProvider, self).__init__(obj_path)
def get_random_embedding_iter(self, batch_size, embedding_ids):
while True:
# np.random.shuffle(self.data.examples)
rand_iter = super(NeverEndingLoopingProvider, self) \
.get_random_embedding_iter(batch_size, embedding_ids)
for labels, images in rand_iter:
yield labels, images
if __name__ == '__main__':
from PIL import Image
pkl_images = PickledImageProvider("../binary/train.obj")
examples = pkl_images.examples
print(len(examples))
b_img0 = examples[0][1] # idx, binary
img0 = bytes_to_file(b_img0)
img_A, img_B = read_split_image(img0)
img = Image.fromarray(np.uint8(img_A), "RGB")
# img.save('my.png')
# mat = misc.imread(img0).astype(np.float)
# side = int(mat.shape[1] / 2)
# assert side * 2 == mat.shape[1]
# img_A = mat[:, :side] # target
# img = Image.fromarray(np.uint8(img_B))
img.show()