File size: 20,522 Bytes
032e687 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 |
# Copyright (c) Facebook, Inc. and its affiliates.
# Modified by Bowen Cheng from https://github.com/sukjunhwang/IFC
import numpy as np
import logging
import sys
from fvcore.transforms.transform import (
HFlipTransform,
NoOpTransform,
VFlipTransform,
BlendTransform,
CropTransform,
PadTransform,
Transform,
TransformList,
)
from PIL import Image
from .detectron2.data2 import transforms as T
class RandomApplyClip(T.Augmentation):
"""
Randomly apply an augmentation with a given probability.
"""
def __init__(self, tfm_or_aug, prob=0.5, clip_frame_cnt=1):
"""
Args:
tfm_or_aug (Transform, Augmentation): the transform or augmentation
to be applied. It can either be a `Transform` or `Augmentation`
instance.
prob (float): probability between 0.0 and 1.0 that
the wrapper transformation is applied
"""
super().__init__()
self.aug = T.augmentation._transform_to_aug(tfm_or_aug)
assert 0.0 <= prob <= 1.0, f"Probablity must be between 0.0 and 1.0 (given: {prob})"
self.prob = prob
self._cnt = 0
self.clip_frame_cnt = clip_frame_cnt
def get_transform(self, *args):
if self._cnt % self.clip_frame_cnt == 0:
self.do = self._rand_range() < self.prob
self._cnt = 0 # avoiding overflow
self._cnt += 1
if self.do:
return self.aug.get_transform(*args)
else:
return NoOpTransform()
def __call__(self, aug_input):
if self._cnt % self.clip_frame_cnt == 0:
self.do = self._rand_range() < self.prob
self._cnt = 0 # avoiding overflow
self._cnt += 1
if self.do:
return self.aug(aug_input)
else:
return NoOpTransform()
class RandomRotationClip(T.Augmentation):
"""
This method returns a copy of this image, rotated the given
number of degrees counter clockwise around the given center.
"""
def __init__(self, angle, prob=0.5, expand=True, center=None, interp=None, clip_frame_cnt=1):
"""
Args:
angle (list[float]): If ``sample_style=="range"``,
a [min, max] interval from which to sample the angle (in degrees).
If ``sample_style=="choice"``, a list of angles to sample from
expand (bool): choose if the image should be resized to fit the whole
rotated image (default), or simply cropped
center (list[[float, float]]): If ``sample_style=="range"``,
a [[minx, miny], [maxx, maxy]] relative interval from which to sample the center,
[0, 0] being the top left of the image and [1, 1] the bottom right.
If ``sample_style=="choice"``, a list of centers to sample from
Default: None, which means that the center of rotation is the center of the image
center has no effect if expand=True because it only affects shifting
"""
super().__init__()
if isinstance(angle, (float, int)):
angle = (angle, angle)
if center is not None and isinstance(center[0], (float, int)):
center = (center, center)
self.angle_save = None
self.center_save = None
self._cnt = 0
self._init(locals())
def get_transform(self, image):
h, w = image.shape[:2]
if self._cnt % self.clip_frame_cnt == 0:
center = None
angle = np.random.uniform(self.angle[0], self.angle[1], size=self.clip_frame_cnt)
if self.center is not None:
center = (
np.random.uniform(self.center[0][0], self.center[1][0]),
np.random.uniform(self.center[0][1], self.center[1][1]),
)
angle = np.sort(angle)
if self._rand_range() < self.prob:
angle = angle[::-1]
self.angle_save = angle
self.center_save = center
self._cnt = 0 # avoiding overflow
angle = self.angle_save[self._cnt]
center = self.center_save
self._cnt += 1
if center is not None:
center = (w * center[0], h * center[1]) # Convert to absolute coordinates
if angle % 360 == 0:
return NoOpTransform()
return T.RotationTransform(h, w, angle, expand=self.expand, center=center, interp=self.interp)
class ResizeScaleClip(T.Augmentation):
"""
Takes target size as input and randomly scales the given target size between `min_scale`
and `max_scale`. It then scales the input image such that it fits inside the scaled target
box, keeping the aspect ratio constant.
This implements the resize part of the Google's 'resize_and_crop' data augmentation:
https://github.com/tensorflow/tpu/blob/master/models/official/detection/utils/input_utils.py#L127
"""
def __init__(
self,
min_scale: float,
max_scale: float,
target_height: int,
target_width: int,
interp: int = Image.BILINEAR,
clip_frame_cnt=1,
):
"""
Args:
min_scale: minimum image scale range.
max_scale: maximum image scale range.
target_height: target image height.
target_width: target image width.
interp: image interpolation method.
"""
super().__init__()
self._init(locals())
self._cnt = 0
def _get_resize(self, image: np.ndarray, scale: float):
input_size = image.shape[:2]
# Compute new target size given a scale.
target_size = (self.target_height, self.target_width)
target_scale_size = np.multiply(target_size, scale)
# Compute actual rescaling applied to input image and output size.
output_scale = np.minimum(
target_scale_size[0] / input_size[0], target_scale_size[1] / input_size[1]
)
output_size = np.round(np.multiply(input_size, output_scale)).astype(int)
return T.ResizeTransform(
input_size[0], input_size[1], output_size[0], output_size[1], self.interp
)
def get_transform(self, image: np.ndarray):
if self._cnt % self.clip_frame_cnt == 0:
random_scale = np.random.uniform(self.min_scale, self.max_scale)
self.random_scale_save = random_scale
self._cnt = 0 # avoiding overflow
self._cnt += 1
random_scale = self.random_scale_save
return self._get_resize(image, random_scale)
class RandomCropClip(T.Augmentation):
"""
Randomly crop a rectangle region out of an image.
"""
def __init__(self, crop_type: str, crop_size, clip_length):
"""
Args:
crop_type (str): one of "relative_range", "relative", "absolute", "absolute_range".
crop_size (tuple[float, float]): two floats, explained below.
- "relative": crop a (H * crop_size[0], W * crop_size[1]) region from an input image of
size (H, W). crop size should be in (0, 1]
- "relative_range": uniformly sample two values from [crop_size[0], 1]
and [crop_size[1]], 1], and use them as in "relative" crop type.
- "absolute" crop a (crop_size[0], crop_size[1]) region from input image.
crop_size must be smaller than the input image size.
- "absolute_range", for an input of size (H, W), uniformly sample H_crop in
[crop_size[0], min(H, crop_size[1])] and W_crop in [crop_size[0], min(W, crop_size[1])].
Then crop a region (H_crop, W_crop).
"""
# TODO style of relative_range and absolute_range are not consistent:
# one takes (h, w) but another takes (min, max)
super().__init__()
self.clip_length = clip_length
self._cnt = 0
self.transform_temp = None
assert crop_type in ["relative_range", "relative", "absolute", "absolute_range"]
self._init(locals())
def get_transform(self, image):
if self._cnt % self.clip_length == 0:
h, w = image.shape[:2]
croph, cropw = self.get_crop_size((h, w))
assert h >= croph and w >= cropw, "Shape computation in {} has bugs.".format(self)
h0 = np.random.randint(h - croph + 1)
w0 = np.random.randint(w - cropw + 1)
self.transform_temp = CropTransform(w0, h0, cropw, croph)
self._cnt = 0
self._cnt += 1
return self.transform_temp
else:
self._cnt += 1
return self.transform_temp
def get_crop_size(self, image_size):
"""
Args:
image_size (tuple): height, width
Returns:
crop_size (tuple): height, width in absolute pixels
"""
h, w = image_size
if self.crop_type == "relative":
ch, cw = self.crop_size
return int(h * ch + 0.5), int(w * cw + 0.5)
elif self.crop_type == "relative_range":
crop_size = np.asarray(self.crop_size, dtype=np.float32)
ch, cw = crop_size + np.random.rand(2) * (1 - crop_size)
return int(h * ch + 0.5), int(w * cw + 0.5)
elif self.crop_type == "absolute":
return (min(self.crop_size[0], h), min(self.crop_size[1], w))
elif self.crop_type == "absolute_range":
assert self.crop_size[0] <= self.crop_size[1]
ch = np.random.randint(min(h, self.crop_size[0]), min(h, self.crop_size[1]) + 1)
cw = np.random.randint(min(w, self.crop_size[0]), min(w, self.crop_size[1]) + 1)
return ch, cw
else:
raise NotImplementedError("Unknown crop type {}".format(self.crop_type))
class FixedSizeCropClip(T.Augmentation):
"""
If `crop_size` is smaller than the input image size, then it uses a random crop of
the crop size. If `crop_size` is larger than the input image size, then it pads
the right and the bottom of the image to the crop size if `pad` is True, otherwise
it returns the smaller image.
"""
def __init__(self, crop_size, pad=True, pad_value=128.0, clip_frame_cnt=1):
"""
Args:
crop_size: target image (height, width).
pad: if True, will pad images smaller than `crop_size` up to `crop_size`
pad_value: the padding value.
"""
super().__init__()
self._init(locals())
self._cnt = 0
def _get_crop(self, image: np.ndarray):
# Compute the image scale and scaled size.
input_size = image.shape[:2]
output_size = self.crop_size
# Add random crop if the image is scaled up.
max_offset = np.subtract(input_size, output_size)
max_offset = np.maximum(max_offset, 0)
if self._cnt % self.clip_frame_cnt == 0:
offset = np.multiply(max_offset, np.random.uniform(0.0, 1.0))
offset = np.round(offset).astype(int)
self.offset_save = offset
self._cnt = 0 # avoiding overflow
self._cnt += 1
offset = self.offset_save
return CropTransform(
offset[1], offset[0], output_size[1], output_size[0], input_size[1], input_size[0]
)
def _get_pad(self, image: np.ndarray):
# Compute the image scale and scaled size.
input_size = image.shape[:2]
output_size = self.crop_size
# Add padding if the image is scaled down.
pad_size = np.subtract(output_size, input_size)
pad_size = np.maximum(pad_size, 0)
original_size = np.minimum(input_size, output_size)
return PadTransform(
0, 0, pad_size[1], pad_size[0], original_size[1], original_size[0], self.pad_value
)
def get_transform(self, image: np.ndarray):
transforms = [self._get_crop(image)]
if self.pad:
transforms.append(self._get_pad(image))
return TransformList(transforms)
class ResizeShortestEdge(T.Augmentation):
"""
Scale the shorter edge to the given size, with a limit of `max_size` on the longer edge.
If `max_size` is reached, then downscale so that the longer edge does not exceed max_size.
"""
def __init__(
self, short_edge_length, max_size=sys.maxsize, sample_style="range", interp=Image.BILINEAR, clip_frame_cnt=1
):
"""
Args:
short_edge_length (list[int]): If ``sample_style=="range"``,
a [min, max] interval from which to sample the shortest edge length.
If ``sample_style=="choice"``, a list of shortest edge lengths to sample from.
max_size (int): maximum allowed longest edge length.
sample_style (str): either "range" or "choice".
"""
super().__init__()
assert sample_style in ["range", "choice", "range_by_clip", "choice_by_clip"], sample_style
self.is_range = ("range" in sample_style)
if isinstance(short_edge_length, int):
short_edge_length = (short_edge_length, short_edge_length)
if self.is_range:
assert len(short_edge_length) == 2, (
"short_edge_length must be two values using 'range' sample style."
f" Got {short_edge_length}!"
)
self._cnt = 0
self._init(locals())
def get_transform(self, image):
if self._cnt % self.clip_frame_cnt == 0:
if self.is_range:
self.size = np.random.randint(self.short_edge_length[0], self.short_edge_length[1] + 1)
else:
self.size = np.random.choice(self.short_edge_length)
if self.size == 0:
return NoOpTransform()
self._cnt = 0 # avoiding overflow
self._cnt += 1
h, w = image.shape[:2]
scale = self.size * 1.0 / min(h, w)
if h < w:
newh, neww = self.size, scale * w
else:
newh, neww = scale * h, self.size
if max(newh, neww) > self.max_size:
scale = self.max_size * 1.0 / max(newh, neww)
newh = newh * scale
neww = neww * scale
neww = int(neww + 0.5)
newh = int(newh + 0.5)
return T.ResizeTransform(h, w, newh, neww, self.interp)
class RandomFlip(T.Augmentation):
"""
Flip the image horizontally or vertically with the given probability.
"""
def __init__(self, prob=0.5, *, horizontal=True, vertical=False, clip_frame_cnt=1):
"""
Args:
prob (float): probability of flip.
horizontal (boolean): whether to apply horizontal flipping
vertical (boolean): whether to apply vertical flipping
"""
super().__init__()
if horizontal and vertical:
raise ValueError("Cannot do both horiz and vert. Please use two Flip instead.")
if not horizontal and not vertical:
raise ValueError("At least one of horiz or vert has to be True!")
self._cnt = 0
self._init(locals())
def get_transform(self, image):
if self._cnt % self.clip_frame_cnt == 0:
self.do = self._rand_range() < self.prob
self._cnt = 0 # avoiding overflow
self._cnt += 1
h, w = image.shape[:2]
if self.do:
if self.horizontal:
return HFlipTransform(w)
elif self.vertical:
return VFlipTransform(h)
else:
return NoOpTransform()
def build_augmentation(cfg, is_train):
logger = logging.getLogger(__name__)
aug_list = []
if is_train:
min_size = cfg.INPUT.MIN_SIZE_TRAIN
max_size = cfg.INPUT.MAX_SIZE_TRAIN
sample_style = cfg.INPUT.MIN_SIZE_TRAIN_SAMPLING
clip_frame_cnt = cfg.INPUT.SAMPLING_FRAME_NUM if "by_clip" in cfg.INPUT.MIN_SIZE_TRAIN_SAMPLING else 1
# Crop
if cfg.INPUT.CROP.ENABLED:
crop_aug = RandomApplyClip(
T.AugmentationList([
ResizeShortestEdge([400, 500, 600], 1333, sample_style, clip_frame_cnt=clip_frame_cnt),
RandomCropClip(cfg.INPUT.PSEUDO.CROP.TYPE, cfg.INPUT.PSEUDO.CROP.SIZE, clip_length=clip_frame_cnt)
]),
clip_frame_cnt=clip_frame_cnt
)
aug_list.append(crop_aug)
# Resize
aug_list.append(ResizeShortestEdge(min_size, max_size, sample_style, clip_frame_cnt=clip_frame_cnt))
# Flip
if cfg.INPUT.RANDOM_FLIP != "none":
if cfg.INPUT.RANDOM_FLIP == "flip_by_clip":
flip_clip_frame_cnt = cfg.INPUT.SAMPLING_FRAME_NUM
else:
flip_clip_frame_cnt = 1
aug_list.append(
# NOTE using RandomFlip modified for the support of flip maintenance
RandomFlip(
horizontal=(cfg.INPUT.RANDOM_FLIP == "horizontal") or (cfg.INPUT.RANDOM_FLIP == "flip_by_clip"),
vertical=cfg.INPUT.RANDOM_FLIP == "vertical",
clip_frame_cnt=flip_clip_frame_cnt,
)
)
# Additional augmentations : brightness, contrast, saturation, rotation
augmentations = cfg.INPUT.AUGMENTATIONS
if "brightness" in augmentations:
aug_list.append(T.RandomBrightness(0.9, 1.1))
if "contrast" in augmentations:
aug_list.append(T.RandomContrast(0.9, 1.1))
if "saturation" in augmentations:
aug_list.append(T.RandomSaturation(0.9, 1.1))
if "rotation" in augmentations:
aug_list.append(
T.RandomRotation(
[-15, 15], expand=False, center=[(0.4, 0.4), (0.6, 0.6)], sample_style="range"
)
)
else:
# Resize
min_size = cfg.INPUT.MIN_SIZE_TEST
max_size = cfg.INPUT.MAX_SIZE_TEST
sample_style = "choice"
aug_list.append(T.ResizeShortestEdge(min_size, max_size, sample_style))
return aug_list
def build_pseudo_augmentation(is_train, force_image_size=1024):
aug_list = []
if is_train:
min_size = (512, 544, 576, 608, 640, 672, 704, 736, 768, 800, 832, 864)
max_size = force_image_size
sample_style = "choice_by_clip"
clip_frame_cnt = 2
random_flip = "flip_by_clip"
# Crop
if True:
crop_aug = RandomApplyClip(
T.AugmentationList([
ResizeShortestEdge(min_size, max_size, sample_style, clip_frame_cnt=clip_frame_cnt),
RandomCropClip("absolute_range", (480, 864), clip_length=clip_frame_cnt),
]),
clip_frame_cnt=clip_frame_cnt
)
aug_list.append(crop_aug)
# Resize
aug_list.append(ResizeShortestEdge(max_size, max_size, sample_style, clip_frame_cnt=clip_frame_cnt))
# Flip
aug_list.append(
# NOTE using RandomFlip modified for the support of flip maintenance
RandomFlip(
horizontal=(random_flip == "horizontal") or (random_flip == "flip_by_clip"),
vertical=random_flip == "vertical",
clip_frame_cnt=clip_frame_cnt,
)
)
# Additional augmentations : brightness, contrast, saturation, rotation
augmentations = ['rotation']
if "brightness" in augmentations:
aug_list.append(T.RandomBrightness(0.9, 1.1))
if "contrast" in augmentations:
aug_list.append(T.RandomContrast(0.9, 1.1))
if "saturation" in augmentations:
aug_list.append(T.RandomSaturation(0.9, 1.1))
if "rotation" in augmentations:
aug_list.append(
RandomRotationClip(
[-15, 15], expand=False, center=[(0.4, 0.4), (0.6, 0.6)], clip_frame_cnt=clip_frame_cnt,
)
)
aug_list.append(
FixedSizeCropClip(
crop_size=(max_size, max_size), clip_frame_cnt=clip_frame_cnt,
)
)
else:
# Resize
min_size = (320, 352, 392, 416, 448, 480, 512, 544, 576, 608, 640, 672, 704, 736, 768, )
max_size = 1024
sample_style = "choice"
aug_list.append(T.ResizeShortestEdge(min_size, max_size, sample_style))
return aug_list |