import torch
from torchvision import transforms
import cv2
import numpy as np
import types
from numpy import random
[docs]def intersect(box_a, box_b):
max_xy = np.minimum(box_a[:, 2:], box_b[2:])
min_xy = np.maximum(box_a[:, :2], box_b[:2])
inter = np.clip((max_xy - min_xy), a_min=0, a_max=np.inf)
return inter[:, 0] * inter[:, 1]
[docs]def jaccard_numpy(box_a, box_b):
"""Compute the jaccard overlap of two sets of boxes. The jaccard overlap
is simply the intersection over union of two boxes.
E.g.: A ∩ B / A ∪ B = A ∩ B / (area(A) + area(B) - A ∩ B)
Args:
box_a: Multiple bounding boxes, Shape: [num_boxes,4]
box_b: Single bounding box, Shape: [4]
Return:
jaccard overlap: Shape: [box_a.shape[0], box_a.shape[1]]
"""
inter = intersect(box_a, box_b)
area_a = ((box_a[:, 2]-box_a[:, 0]) *
(box_a[:, 3]-box_a[:, 1])) # [A,B]
area_b = ((box_b[2]-box_b[0]) *
(box_b[3]-box_b[1])) # [A,B]
union = area_a + area_b - inter
return inter / union # [A,B]
[docs]class Compose(object):
"""Composes several augmentations together.
Args:
transforms (List[Transform]): list of transforms to compose.
Example:
>>> augmentations.Compose([
>>> transforms.CenterCrop(10),
>>> transforms.ToTensor(),
>>> ])
"""
def __init__(self, transforms):
self.transforms = transforms
def __call__(self, img, boxes=None, labels=None):
for t in self.transforms:
img, boxes, labels = t(img, boxes, labels)
return img, boxes, labels
[docs]class Lambda(object):
"""Applies a lambda as a transform."""
def __init__(self, lambd):
assert isinstance(lambd, types.LambdaType)
self.lambd = lambd
def __call__(self, img, boxes=None, labels=None):
return self.lambd(img, boxes, labels)
[docs]class ConvertFromInts(object):
def __call__(self, image, boxes=None, labels=None):
return image.astype(np.float32), boxes, labels
[docs]class ToAbsoluteCoords(object):
def __call__(self, image, boxes=None, labels=None):
height, width, channels = image.shape
boxes[:, 0] *= width
boxes[:, 2] *= width
boxes[:, 1] *= height
boxes[:, 3] *= height
return image, boxes, labels
[docs]class ToPercentCoords(object):
def __call__(self, image, boxes=None, labels=None):
height, width, channels = image.shape
boxes[:, 0] /= width
boxes[:, 2] /= width
boxes[:, 1] /= height
boxes[:, 3] /= height
return image, boxes, labels
[docs]class Resize(object):
def __init__(self, size:tuple):
self.size = size
def __call__(self, image, boxes=None, labels=None):
image = cv2.resize(image, self.size)
return image, boxes, labels
[docs]class RandomSaturation(object):
def __init__(self, lower=0.5, upper=1.5):
self.lower = lower
self.upper = upper
assert self.upper >= self.lower, "contrast upper must be >= lower."
assert self.lower >= 0, "contrast lower must be non-negative."
def __call__(self, image, boxes=None, labels=None):
if random.randint(2):
image[:, :, 1] *= random.uniform(self.lower, self.upper)
return image, boxes, labels
[docs]class RandomHue(object):
def __init__(self, delta=18.0):
assert delta >= 0.0 and delta <= 360.0
self.delta = delta
def __call__(self, image, boxes=None, labels=None):
if random.randint(2):
image[:, :, 0] += random.uniform(-self.delta, self.delta)
image[:, :, 0][image[:, :, 0] > 360.0] -= 360.0
image[:, :, 0][image[:, :, 0] < 0.0] += 360.0
return image, boxes, labels
[docs]class RandomLightingNoise(object):
def __init__(self):
self.perms = ((0, 1, 2), (0, 2, 1),
(1, 0, 2), (1, 2, 0),
(2, 0, 1), (2, 1, 0))
def __call__(self, image, boxes=None, labels=None):
if random.randint(2):
swap = self.perms[random.randint(len(self.perms))]
shuffle = SwapChannels(swap) # shuffle channels
image = shuffle(image)
return image, boxes, labels
[docs]class ConvertColor(object):
def __init__(self, current='BGR', transform='HSV'):
self.transform = transform
self.current = current
def __call__(self, image, boxes=None, labels=None):
if self.current == 'BGR' and self.transform == 'HSV':
image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
elif self.current == 'HSV' and self.transform == 'BGR':
image = cv2.cvtColor(image, cv2.COLOR_HSV2BGR)
else:
raise NotImplementedError
return image, boxes, labels
[docs]class RandomContrast(object):
def __init__(self, lower=0.5, upper=1.5):
self.lower = lower
self.upper = upper
assert self.upper >= self.lower, "contrast upper must be >= lower."
assert self.lower >= 0, "contrast lower must be non-negative."
# expects float image
def __call__(self, image, boxes=None, labels=None):
if random.randint(2):
alpha = random.uniform(self.lower, self.upper)
image *= alpha
return image, boxes, labels
[docs]class RandomBrightness(object):
def __init__(self, delta=32):
assert delta >= 0.0
assert delta <= 255.0
self.delta = delta
def __call__(self, image, boxes=None, labels=None):
if random.randint(2):
delta = random.uniform(-self.delta, self.delta)
image += delta
return image, boxes, labels
[docs]class ToCV2Image(object):
def __call__(self, tensor, boxes=None, labels=None):
return tensor.cpu().numpy().astype(np.float32).transpose((1, 2, 0)), boxes, labels
[docs]class ToTensor(object):
def __call__(self, cvimage, boxes=None, labels=None):
return torch.from_numpy(cvimage.astype(np.float32)).permute(2, 0, 1), boxes, labels
[docs]class RandomSampleCrop(object):
"""Crop
Arguments:
img (Image): the image being input during training
boxes (Tensor): the original bounding boxes in pt form
labels (Tensor): the class labels for each bbox
mode (float tuple): the min and max jaccard overlaps
Return:
(img, boxes, classes)
img (Image): the cropped image
boxes (Tensor): the adjusted bounding boxes in pt form
labels (Tensor): the class labels for each bbox
"""
def __init__(self, scale=(0.3,1.), ratio=(0.5,2), num_attempts=50):
self.scale = scale
self.ratio = ratio
self.num_attempts = num_attempts
self.sample_options = (
# using entire original input image
None,
# sample a patch s.t. MIN jaccard w/ obj in .1,.3,.5,.7,.9
(0.1, None),
(0.3, None),
(0.5, None),
(0.7, None),
(0.9, None),
# randomly sample a patch
(None, None),
)
def __call__(self, image, boxes=None, labels=None):
height, width, _ = image.shape
while True:
# randomly choose a mode
mode = random.choice(self.sample_options)
if mode is None:
return image, boxes, labels
min_iou, max_iou = mode
if min_iou is None:
min_iou = float('-inf')
if max_iou is None:
max_iou = float('inf')
# max trails (50)
for _ in range(self.num_attempts):
current_image = image
w = random.uniform(self.scale[0] * width, self.scale[1] * width)
h = random.uniform(self.scale[0] * height, self.scale[1] * height)
# aspect ratio constraint b/t .5 & 2
if h / w < self.ratio[1] or h / w > self.ratio[0]:
continue
left = random.uniform(width - w)
top = random.uniform(height - h)
# convert to integer rect x1,y1,x2,y2
rect = np.array([int(left), int(top), int(left+w), int(top+h)])
# calculate IoU (jaccard overlap) b/t the cropped and gt boxes
overlap = jaccard_numpy(boxes, rect)
# is min and max overlap constraint satisfied? if not try again
if overlap.min() < min_iou and max_iou < overlap.max():
continue
# cut the crop from the image
current_image = current_image[rect[1]:rect[3], rect[0]:rect[2],
:]
# keep overlap with gt box IF center in sampled patch
centers = (boxes[:, :2] + boxes[:, 2:]) / 2.0
# mask in all gt boxes that above and to the left of centers
m1 = (rect[0] < centers[:, 0]) * (rect[1] < centers[:, 1])
# mask in all gt boxes that under and to the right of centers
m2 = (rect[2] > centers[:, 0]) * (rect[3] > centers[:, 1])
# mask in that both m1 and m2 are true
mask = m1 * m2
# have any valid boxes? try again if not
if not mask.any():
continue
# take only matching gt boxes
current_boxes = boxes[mask, :].copy()
# take only matching gt labels
current_labels = labels[mask]
# should we use the box left and top corner or the crop's
current_boxes[:, :2] = np.maximum(current_boxes[:, :2],
rect[:2])
# adjust to crop (by substracting crop's left,top)
current_boxes[:, :2] -= rect[:2]
current_boxes[:, 2:] = np.minimum(current_boxes[:, 2:],
rect[2:])
# adjust to crop (by substracting crop's left,top)
current_boxes[:, 2:] -= rect[:2]
return current_image, current_boxes, current_labels
[docs]class Expand(object):
def __init__(self, mean=0, max_expand_ratio=2):
"""
Args:
max_expand_ratio: Ratio of canvas size to input size, must be > 1.
"""
self.mean = mean
self.max_expand_ratio = max_expand_ratio
def __call__(self, image, boxes, labels):
if random.randint(2):
return image, boxes, labels
height, width, depth = image.shape
ratio = random.uniform(1, self.max_expand_ratio)
left = random.uniform(0, width*ratio - width)
top = random.uniform(0, height*ratio - height)
expand_image = np.zeros(
(int(height*ratio), int(width*ratio), depth),
dtype=image.dtype)
expand_image[:, :, :] = self.mean
expand_image[int(top):int(top + height),
int(left):int(left + width)] = image
image = expand_image
boxes = boxes.copy()
boxes[:, :2] += (int(left), int(top))
boxes[:, 2:] += (int(left), int(top))
return image, boxes, labels
[docs]class RandomMirror(object):
def __call__(self, image, boxes, classes):
_, width, _ = image.shape
if random.randint(2):
image = image[:, ::-1]
boxes = boxes.copy()
boxes[:, 0::2] = width - boxes[:, 2::-2]
return image, boxes, classes
[docs]class SwapChannels(object):
"""Transforms a tensorized image by swapping the channels in the order
specified in the swap tuple.
Args:
swaps (int triple): final order of channels, (2, 1, 0)
"""
def __init__(self, swaps):
self.swaps = swaps
def __call__(self, image):
"""
Args:
image (Tensor): image tensor to be transformed
Returns:
a tensor with channels swapped according to swap
"""
# if torch.is_tensor(image):
# image = image.data.cpu().numpy()
# else:
# image = np.array(image)
image = image[:, :, self.swaps]
return image
[docs]class PhotometricDistort(object):
def __init__(self, hue_delta=18., bri_delta=32,
contrast_range=(0.5,1.5),
saturation_range=(0.5,1.5)):
self.pd = [
RandomContrast(*contrast_range),
# temp remove
# ConvertColor(transform='HSV'),
# RandomSaturation(*saturation_range),
# RandomHue(delta=hue_delta),
# ConvertColor(current='HSV', transform='BGR'),
# RandomContrast(*contrast_range)
]
self.rand_brightness = RandomBrightness(delta=bri_delta)
# self.rand_light_noise = RandomLightingNoise()
def __call__(self, image, boxes, labels):
im = image.copy()
im, boxes, labels = self.rand_brightness(im, boxes, labels)
if random.randint(2):
distort = Compose(self.pd[:-1])
else:
distort = Compose(self.pd[1:])
im, boxes, labels = distort(im, boxes, labels)
# return self.rand_light_noise(im, boxes, labels)
return im, boxes, labels
[docs]class ToGPU(object):
def __call__(self, img, boxes, labels):
return img.cuda(), boxes, labels
[docs]class Normalize(object):
def __init__(self, mean=0, std=255):
self.mean = mean
self.std = std
def __call__(self, img, boxes, labels):
img = (img - self.mean) / self.std
return img, boxes, labels
[docs]class ToXYWH(object):
def __call__(self, img, boxes, labels):
boxes[:,2:] -= boxes[:,:2]
return img, boxes, labels