def preprocess_image(image, desired_size, stride):
image = input_utils.normalize_image(image)
image, image_info = input_utils.resize_and_crop_image(
image,
desired_size,
padded_size=input_utils.compute_padded_size(desired_size, stride))
return image, image_info
def parse_single_example(serialized_example, params):
"""Parses a singel serialized TFExample string."""
if 'retinanet_parser' in dir(params):
parser_params = params.retinanet_parser
decoder = tf_example_decoder.TfExampleDecoder()
else:
parser_params = params.maskrcnn_parser
decoder = tf_example_decoder.TfExampleDecoder(include_mask=True)
data = decoder.decode(serialized_example)
image = data['image']
source_id = data['source_id']
source_id = dataloader_utils.process_source_id(source_id)
height = data['height']
width = data['width']
boxes = data['groundtruth_boxes']
boxes = box_utils.denormalize_boxes(boxes, tf.shape(image)[:2])
classes = data['groundtruth_classes']
is_crowds = data['groundtruth_is_crowd']
areas = data['groundtruth_area']
masks = data.get('groundtruth_instance_masks_png', None)
image = input_utils.normalize_image(image)
image, image_info = input_utils.resize_and_crop_image(
image,
parser_params.output_size,
padded_size=input_utils.compute_padded_size(
parser_params.output_size,
2 ** params.architecture.max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
labels = {
'image_info': image_info,
}
groundtruths = {
'source_id': source_id,
'height': height,
'width': width,
'num_detections': tf.shape(classes),
'boxes': boxes,
'classes': classes,
'areas': areas,
'is_crowds': tf.cast(is_crowds, tf.int32),
}
if masks is not None:
groundtruths['masks'] = masks
return image, labels, groundtruths
def parse_single_example(serialized_example, params):
"""Parses a singel serialized TFExample string."""
decoder = tf_example_decoder.TfExampleDecoder()
data = decoder.decode(serialized_example)
image = data['image']
source_id = data['source_id']
source_id = dataloader_utils.process_source_id(source_id)
height = data['height']
width = data['width']
boxes = data['groundtruth_boxes']
boxes = box_utils.denormalize_boxes(boxes, tf.shape(image)[:2])
classes = data['groundtruth_classes']
is_crowds = data['groundtruth_is_crowd']
areas = data['groundtruth_area']
image = input_utils.normalize_image(image)
image, image_info = input_utils.resize_and_crop_image(
image,
params.retinanet_parser.output_size,
padded_size=input_utils.compute_padded_size(
params.retinanet_parser.output_size, 2**params.anchor.max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
anchors = anchor.Anchor(params.anchor.min_level, params.anchor.max_level,
params.anchor.num_scales,
params.anchor.aspect_ratios,
params.anchor.anchor_size,
image.get_shape().as_list()[:2])
labels = {
'anchor_boxes': anchors.multilevel_boxes,
'image_info': image_info,
}
groundtruths = {
'source_id': source_id,
'height': height,
'width': width,
'num_detections': tf.shape(classes),
'boxes': boxes,
'classes': classes,
'areas': areas,
'is_crowds': tf.cast(is_crowds, tf.int32),
}
return image, labels, groundtruths
def _parse_eval_data(self, data):
"""Parses data for training and evaluation."""
image, label = self._prepare_image_and_label(data)
# The label is first offset by +1 and then padded with 0.
label += 1
label = tf.expand_dims(label, axis=3)
if self._resize_eval:
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image, self._output_size, self._output_size)
# Resizes and crops mask.
image_scale = image_info[2, :]
offset = image_info[3, :]
label = input_utils.resize_and_crop_masks(label, image_scale,
self._output_size, offset)
else:
# Pads image and mask to output size.
image = tf.image.pad_to_bounding_box(image, 0, 0, self._output_size[0],
self._output_size[1])
label = tf.image.pad_to_bounding_box(label, 0, 0, self._output_size[0],
self._output_size[1])
label -= 1
label = tf.where(tf.equal(label, -1),
self._ignore_label * tf.ones_like(label), label)
label = tf.squeeze(label, axis=0)
valid_mask = tf.not_equal(label, self._ignore_label)
labels = {
'masks': label,
'valid_masks': valid_mask
}
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
return image, labels
def _parse_train_data(self, data):
"""Parses data for training and evaluation."""
image, label = self._prepare_image_and_label(data)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, label = input_utils.random_horizontal_flip(image, masks=label)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
self._output_size,
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
# Resizes and crops boxes.
image_scale = image_info[2, :]
offset = image_info[3, :]
# Pad label and make sure the padded region assigned to the ignore label.
# The label is first offset by +1 and then padded with 0.
label += 1
label = tf.expand_dims(label, axis=3)
label = input_utils.resize_and_crop_masks(
label, image_scale, self._output_size, offset)
label -= 1
label = tf.where(tf.equal(label, -1),
self._ignore_label * tf.ones_like(label), label)
label = tf.squeeze(label, axis=0)
valid_mask = tf.not_equal(label, self._ignore_label)
labels = {
'masks': label,
'valid_masks': valid_mask
}
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
return image, labels
def _preprocess(self, image: np.ndarray) -> tuple[tf.Tensor, tf.Tensor]:
image = tf.convert_to_tensor(image, tf.uint8)
image = input_utils.normalize_image(image)
image, image_info = input_utils.resize_and_crop_image(
image=image,
desired_size=self.resize_shape,
padded_size=input_utils.compute_padded_size(
desired_size=self.resize_shape,
stride=2**self.max_level,
),
aug_scale_min=1.0,
aug_scale_max=1.0,
)
# image_info: (4, 2)
# [
# [original_height, original_width],
# [desired_height, desired_width ],
# [y_scale, x_scale ],
# [y_offset, x_offset ]
# ]
image.set_shape([self.resize_shape[0], self.resize_shape[1], 3])
return image, image_info
def parse_predict_data(self, data):
"""Parse data for ShapeMask training."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
masks = data['groundtruth_instance_masks']
# If not using category, makes all categories with id = 0.
if not self._use_category:
classes = tf.cast(tf.greater(classes, 0), dtype=tf.float32)
image = self.get_normalized_image(data)
# Converts boxes from normalized coordinates to pixel coordinates.
image_shape = tf.shape(image)[0:2]
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
self._output_size,
aug_scale_min=1.0,
aug_scale_max=1.0)
image_scale = image_info[2, :]
offset = image_info[3, :]
# Resizes and crops boxes and masks.
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
image_info[1, :], offset)
masks = input_utils.resize_and_crop_masks(
tf.expand_dims(masks, axis=-1), image_scale, self._output_size,
offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size, self._output_size)
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
labels = {
'anchor_boxes': input_anchor.multilevel_boxes,
'image_info': image_info,
}
if self._mode == ModeKeys.PREDICT_WITH_GT:
# Converts boxes from normalized coordinates to pixel coordinates.
groundtruths = {
'source_id':
data['source_id'],
'height':
data['height'],
'width':
data['width'],
'num_detections':
tf.shape(data['groundtruth_classes']),
'boxes':
box_utils.denormalize_boxes(data['groundtruth_boxes'],
image_shape),
'classes':
data['groundtruth_classes'],
# 'masks': tf.squeeze(masks, axis=-1),
'areas':
data['groundtruth_area'],
'is_crowds':
tf.cast(data['groundtruth_is_crowd'], tf.int32),
}
groundtruths['source_id'] = dataloader_utils.process_source_id(
groundtruths['source_id'])
groundtruths = dataloader_utils.pad_groundtruths_to_fixed_size(
groundtruths, self._max_num_instances)
# Computes training labels.
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# Packs labels for model_fn outputs.
labels.update({
'cls_targets': cls_targets,
'box_targets': box_targets,
'num_positives': num_positives,
'groundtruths': groundtruths,
})
return {
'images': image,
'labels': labels,
}
def parse_train_data(self, data):
"""Parse data for ShapeMask training."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
masks = data['groundtruth_instance_masks']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
masks = tf.gather(masks, indices)
# If not using category, makes all categories with id = 0.
if not self._use_category:
classes = tf.cast(tf.greater(classes, 0), dtype=tf.float32)
image = self.get_normalized_image(data)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, boxes, masks = input_utils.random_horizontal_flip(
image, boxes, masks)
# Converts boxes from normalized coordinates to pixel coordinates.
image_shape = tf.shape(image)[0:2]
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
self._output_size,
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
self._train_image_scale = image_info[2, :]
self._train_offset = image_info[3, :]
# Resizes and crops boxes and masks.
boxes = input_utils.resize_and_crop_boxes(boxes,
self._train_image_scale,
image_info[1, :],
self._train_offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
masks = tf.gather(masks, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size, self._output_size)
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# Sample groundtruth masks/boxes/classes for mask branch.
num_masks = tf.shape(masks)[0]
mask_shape = tf.shape(masks)[1:3]
# Pad sampled boxes/masks/classes to a constant batch size.
padded_boxes = input_utils.pad_to_fixed_size(boxes,
self._num_sampled_masks)
padded_classes = input_utils.pad_to_fixed_size(classes,
self._num_sampled_masks)
padded_masks = input_utils.pad_to_fixed_size(masks,
self._num_sampled_masks)
# Randomly sample groundtruth masks for mask branch training. For the image
# without groundtruth masks, it will sample the dummy padded tensors.
rand_indices = tf.random.shuffle(
tf.range(tf.maximum(num_masks, self._num_sampled_masks)))
rand_indices = tf.mod(rand_indices, tf.maximum(num_masks, 1))
rand_indices = rand_indices[0:self._num_sampled_masks]
rand_indices = tf.reshape(rand_indices, [self._num_sampled_masks])
sampled_boxes = tf.gather(padded_boxes, rand_indices)
sampled_classes = tf.gather(padded_classes, rand_indices)
sampled_masks = tf.gather(padded_masks, rand_indices)
# Jitter the sampled boxes to mimic the noisy detections.
sampled_boxes = box_utils.jitter_boxes(
sampled_boxes, noise_scale=self._box_jitter_scale)
sampled_boxes = box_utils.clip_boxes(sampled_boxes, self._output_size)
# Compute mask targets in feature crop. A feature crop fully contains a
# sampled box.
mask_outer_boxes = box_utils.compute_outer_boxes(
sampled_boxes, tf.shape(image)[0:2], scale=self._outer_box_scale)
mask_outer_boxes = box_utils.clip_boxes(mask_outer_boxes,
self._output_size)
# Compensate the offset of mask_outer_boxes to map it back to original image
# scale.
mask_outer_boxes_ori = mask_outer_boxes
mask_outer_boxes_ori += tf.tile(
tf.expand_dims(self._train_offset, axis=0), [1, 2])
mask_outer_boxes_ori /= tf.tile(
tf.expand_dims(self._train_image_scale, axis=0), [1, 2])
norm_mask_outer_boxes_ori = box_utils.normalize_boxes(
mask_outer_boxes_ori, mask_shape)
# Set sampled_masks shape to [batch_size, height, width, 1].
sampled_masks = tf.cast(tf.expand_dims(sampled_masks, axis=-1),
tf.float32)
mask_targets = tf.image.crop_and_resize(
sampled_masks,
norm_mask_outer_boxes_ori,
box_ind=tf.range(self._num_sampled_masks),
crop_size=[self._mask_crop_size, self._mask_crop_size],
method='bilinear',
extrapolation_value=0,
name='train_mask_targets')
mask_targets = tf.where(tf.greater_equal(mask_targets, 0.5),
tf.ones_like(mask_targets),
tf.zeros_like(mask_targets))
mask_targets = tf.squeeze(mask_targets, axis=-1)
if self._up_sample_factor > 1:
fine_mask_targets = tf.image.crop_and_resize(
sampled_masks,
norm_mask_outer_boxes_ori,
box_ind=tf.range(self._num_sampled_masks),
crop_size=[
self._mask_crop_size * self._up_sample_factor,
self._mask_crop_size * self._up_sample_factor
],
method='bilinear',
extrapolation_value=0,
name='train_mask_targets')
fine_mask_targets = tf.where(
tf.greater_equal(fine_mask_targets, 0.5),
tf.ones_like(fine_mask_targets),
tf.zeros_like(fine_mask_targets))
fine_mask_targets = tf.squeeze(fine_mask_targets, axis=-1)
else:
fine_mask_targets = mask_targets
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
valid_image = tf.cast(tf.not_equal(num_masks, 0), tf.int32)
if self._mask_train_class == 'all':
mask_is_valid = valid_image * tf.ones_like(sampled_classes,
tf.int32)
else:
# Get the intersection of sampled classes with training splits.
mask_valid_classes = tf.cast(
tf.expand_dims(
class_utils.coco_split_class_ids(self._mask_train_class),
1), sampled_classes.dtype)
match = tf.reduce_any(
tf.equal(tf.expand_dims(sampled_classes, 0),
mask_valid_classes), 0)
mask_is_valid = valid_image * tf.cast(match, tf.int32)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
# For ShapeMask.
'mask_boxes': sampled_boxes,
'mask_outer_boxes': mask_outer_boxes,
'mask_targets': mask_targets,
'fine_mask_targets': fine_mask_targets,
'mask_classes': sampled_classes,
'mask_is_valid': mask_is_valid,
}
return image, labels
def _parse_train_data(self, data):
"""Parses data for training.
Args:
data: the decoded tensor dictionary from TfExampleDecoder.
Returns:
image: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
labels: a dictionary of tensors used for training. The following describes
{key: value} pairs in the dictionary.
image: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
image_info: a 2D `Tensor` that encodes the information of the image and
the applied preprocessing. It is in the format of
[[original_height, original_width], [scaled_height, scaled_width],
num_groundtrtuhs: number of objects.
boxes: Groundtruth bounding box annotations. The box is represented
in [y1, x1, y2, x2] format. The coordinates are w.r.t the scaled
image that is fed to the network. The tennsor is padded with -1 to
the fixed dimension [self._max_num_instances, 4].
classes: Groundtruth classes annotations. The tennsor is padded
with -1 to the fixed dimension [self._max_num_instances].
masks: groundtrugh masks cropped by the bounding box and
resized to a fixed size determined by mask_crop_size.
pasted_objects_mask: a binary tensor with the same size as image which
is computed as the union of all the objects masks.
"""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
if self._include_mask:
masks = data['groundtruth_instance_masks']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
if self._include_mask:
masks = tf.gather(masks, indices)
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
if self._include_mask:
image, boxes, masks = input_utils.random_horizontal_flip(
image, boxes, masks)
else:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Converts boxes from normalized coordinates to pixel coordinates.
# Now the coordinates of boxes are w.r.t. the original image.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
# Resizes and crops boxes.
# Now the coordinates of boxes are w.r.t the scaled image.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
image_info[1, :], offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
if self._include_mask:
masks = tf.gather(masks, indices)
uncropped_masks = tf.cast(masks, tf.int8)
uncropped_masks = tf.expand_dims(uncropped_masks, axis=3)
uncropped_masks = input_utils.resize_and_crop_masks(
uncropped_masks, image_scale, self._output_size, offset)
# Transfer boxes to the original image space and do normalization.
cropped_boxes = boxes + tf.tile(tf.expand_dims(offset, axis=0),
[1, 2])
cropped_boxes /= tf.tile(tf.expand_dims(image_scale, axis=0),
[1, 2])
cropped_boxes = box_utils.normalize_boxes(cropped_boxes,
image_shape)
num_masks = tf.shape(masks)[0]
masks = tf.image.crop_and_resize(
tf.expand_dims(masks, axis=-1),
cropped_boxes,
box_indices=tf.range(num_masks, dtype=tf.int32),
crop_size=[self._mask_crop_size, self._mask_crop_size],
method='bilinear')
masks = tf.squeeze(masks, axis=-1)
indices = tf.range(start=0, limit=tf.shape(classes)[0], dtype=tf.int32)
# Samples the numbers of masks for pasting.
m = tf.random.uniform(shape=[],
maxval=tf.shape(classes)[0] + 1,
dtype=tf.int32)
m = tf.math.minimum(m, tf.shape(classes)[0])
# Shuffles the indices of objects and keep the first m objects for pasting.
shuffled_indices = tf.random.shuffle(indices)
shuffled_indices = tf.slice(shuffled_indices, [0], [m])
boxes = tf.gather(boxes, shuffled_indices)
masks = tf.gather(masks, shuffled_indices)
classes = tf.gather(classes, shuffled_indices)
uncropped_masks = tf.gather(uncropped_masks, shuffled_indices)
pasted_objects_mask = tf.reduce_max(uncropped_masks, 0)
pasted_objects_mask = tf.cast(pasted_objects_mask, tf.bool)
labels = {
'image': image,
'image_info': image_info,
'num_groundtrtuhs': tf.shape(classes)[0],
'boxes': boxes,
'masks': masks,
'classes': classes,
'pasted_objects_mask': pasted_objects_mask,
}
return labels
def _parse_train_data_v2(self, data):
"""Parses data for training.
Args:
data: the decoded tensor dictionary from TfExampleDecoder.
Returns:
image: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
labels: a dictionary of tensors used for training. The following describes
{key: value} pairs in the dictionary.
image_info: a 2D `Tensor` that encodes the information of the image and
the applied preprocessing. It is in the format of
[[original_height, original_width], [scaled_height, scaled_width],
anchor_boxes: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, 4] representing anchor boxes at each level.
rpn_score_targets: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, anchors_per_location]. The height_l and
width_l represent the dimension of class logits at l-th level.
rpn_box_targets: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, anchors_per_location * 4]. The height_l and
width_l represent the dimension of bounding box regression output at
l-th level.
gt_boxes: Groundtruth bounding box annotations. The box is represented
in [y1, x1, y2, x2] format. The coordinates are w.r.t the scaled
image that is fed to the network. The tennsor is padded with -1 to
the fixed dimension [self._max_num_instances, 4].
gt_classes: Groundtruth classes annotations. The tennsor is padded
with -1 to the fixed dimension [self._max_num_instances].
gt_masks: groundtrugh masks cropped by the bounding box and
resized to a fixed size determined by mask_crop_size.
"""
if self._use_autoaugment:
try:
from utils import (
autoaugment_utils, ) # pylint: disable=g-import-not-at-top
except ImportError as e:
logging.exception("Autoaugment is not supported in TF 2.x.")
raise e
classes = data["groundtruth_classes"]
boxes = data["groundtruth_boxes"]
masks = None
attributes = None
if self._include_mask:
masks = data["groundtruth_instance_masks"]
if self._num_attributes:
attributes = data["groundtruth_attributes"]
is_crowds = data["groundtruth_is_crowd"]
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(input=classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
pred=tf.greater(tf.size(input=is_crowds), 0),
true_fn=lambda: tf.compat.v1.where(
tf.logical_not(is_crowds))[:, 0],
false_fn=lambda: tf.cast(tf.range(num_groundtrtuhs), tf.
int64),
)
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
if self._include_mask:
masks = tf.gather(masks, indices)
if self._num_attributes:
attributes = tf.gather(attributes, indices)
# Gets original image and its size.
image = data["image"]
# NOTE: The autoaugment method works best when used alongside the standard
# horizontal flipping of images along with size jittering and normalization.
if self._use_autoaugment and not self._apply_autoaugment_after_resizing:
(
image,
boxes,
masks,
) = autoaugment_utils.distort_image_and_masks_with_autoaugment(
image, boxes, masks, self._autoaugment_policy_name)
# Flips image randomly during training.
if self._aug_rand_hflip:
if self._include_mask:
image, boxes, masks = input_utils.random_horizontal_flip(
image, boxes, masks)
else:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Resizes and crops image.
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max,
)
# Converts boxes from normalized coordinates to pixel coordinates.
# Now the coordinates of boxes are w.r.t. the original image.
orig_image_shape = image_info[0]
boxes = box_utils.denormalize_boxes(boxes, orig_image_shape)
# Resizes and crops boxes.
# Now the coordinates of boxes are w.r.t the scaled image.
rescaled_image_shape = tf.shape(input=image)[:2]
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
rescaled_image_shape, offset)
# Filters out ground truth boxes that are all zeros.
boxes, classes, masks, attributes = self._remove_empty_boxes(
boxes, classes, masks, attributes)
# apply the autoaugment after resizing
if self._use_autoaugment and self._apply_autoaugment_after_resizing:
# prepare image and boxes for the autoaugment
image = tf.image.convert_image_dtype(image, dtype=tf.uint8)
boxes = box_utils.normalize_boxes(boxes, rescaled_image_shape)
# prepare masks for the autoaugment
masks = tf.expand_dims(masks, axis=-1)
scaled_mask_size = tf.cast(
tf.round(orig_image_shape * image_scale), tf.int32)
scaled_masks = tf.image.resize(
masks, scaled_mask_size, method=tf.image.ResizeMethod.BILINEAR)
offset_int = tf.cast(offset, tf.int32)
masks = scaled_masks[:, offset_int[0]:offset_int[0] +
rescaled_image_shape[0],
offset_int[1]:offset_int[1] +
rescaled_image_shape[1], ]
masks = tf.squeeze(masks, axis=-1)
masks = tf.cast(tf.greater(masks, 0.5), tf.float32)
# apply the autoaugment
(
image,
boxes,
masks,
) = autoaugment_utils.distort_image_and_masks_with_autoaugment(
image, boxes, masks, self._autoaugment_policy_name)
# convert the image back to float32 and denormalize bboxes
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
boxes = box_utils.denormalize_boxes(boxes, rescaled_image_shape)
# filters out empty bboxes
boxes, classes, masks, attributes = self._remove_empty_boxes(
boxes, classes, masks, attributes)
if self._include_mask:
if self._use_autoaugment and self._apply_autoaugment_after_resizing:
# don't rescale boxes as masks were already resized
cropped_boxes = box_utils.normalize_boxes(
boxes, rescaled_image_shape)
else:
# transfer boxes to the original image space
cropped_boxes = boxes + tf.tile(tf.expand_dims(offset, axis=0),
[1, 2])
cropped_boxes /= tf.tile(tf.expand_dims(image_scale, axis=0),
[1, 2])
cropped_boxes = box_utils.normalize_boxes(
cropped_boxes, orig_image_shape)
num_masks = tf.shape(input=masks)[0]
masks = tf.image.crop_and_resize(
tf.expand_dims(masks, axis=-1),
cropped_boxes,
box_indices=tf.range(num_masks, dtype=tf.int32),
crop_size=[self._mask_crop_size, self._mask_crop_size],
method="bilinear",
)
masks = tf.squeeze(masks, axis=-1)
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# pad the image
padded_size = input_utils.compute_padded_size(self._output_size,
2**self._max_level)
image = tf.image.pad_to_bounding_box(image, 0, 0, padded_size[0],
padded_size[1])
image_height, image_width, _ = image.get_shape().as_list()
# Assigns anchor targets.
# Note that after the target assignment, box targets are absolute pixel
# offsets w.r.t. the scaled image.
input_anchor = anchor.Anchor(
self._min_level,
self._max_level,
self._num_scales,
self._aspect_ratios,
self._anchor_size,
(image_height, image_width),
)
anchor_labeler = anchor.RpnAnchorLabeler(
input_anchor,
self._rpn_match_threshold,
self._rpn_unmatched_threshold,
self._rpn_batch_size_per_im,
self._rpn_fg_fraction,
)
rpn_score_targets, rpn_box_targets = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=-1), dtype=tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
"anchor_boxes": input_anchor.multilevel_boxes,
"image_info": image_info,
"rpn_score_targets": rpn_score_targets,
"rpn_box_targets": rpn_box_targets,
}
labels["gt_boxes"] = input_utils.pad_to_fixed_size(
boxes, self._max_num_instances, -1)
labels["gt_classes"] = input_utils.pad_to_fixed_size(
classes, self._max_num_instances, -1)
if self._include_mask:
labels["gt_masks"] = input_utils.pad_to_fixed_size(
masks, self._max_num_instances, -1)
if self._num_attributes:
labels["gt_attributes"] = input_utils.pad_to_fixed_size(
attributes, self._max_num_instances, -1)
return image, labels
def _parse_train_data(self, data):
"""Parses data for training and evaluation."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
# Gets original image and its size.
image = data['image']
# NOTE: The autoaugment method works best when used alongside the standard
# horizontal flipping of images along with size jittering and normalization.
if self._use_autoaugment:
try:
from utils import autoaugment_utils # pylint: disable=g-import-not-at-top
except ImportError as e:
logging.exception('Autoaugment is not supported in TF 2.x.')
raise e
image, boxes = autoaugment_utils.distort_image_with_autoaugment(
image, boxes, self._autoaugment_policy_name)
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Converts boxes from normalized coordinates to pixel coordinates.
# Now the coordinates of boxes are w.r.t. the original image.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
image_height, image_width, _ = image.get_shape().as_list()
# Resizes and crops boxes.
# Now the coordinates of boxes are w.r.t the scaled image.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
(image_height, image_width),
offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchor targets.
# Note that after the target assignment, box targets are absolute pixel
# offsets w.r.t. the scaled image.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
}
return image, labels
def _parse_predict_data(self, data):
"""Parses data for prediction.
Args:
data: the decoded tensor dictionary from TfExampleDecoder.
Returns:
A dictionary of {'images': image, 'labels': labels} where
images: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
labels: a dictionary of tensors used for training. The following
describes {key: value} pairs in the dictionary.
source_ids: Source image id. Default value -1 if the source id is
empty in the groundtruth annotation.
image_info: a 2D `Tensor` that encodes the information of the image
and the applied preprocessing. It is in the format of
[[original_height, original_width], [scaled_height, scaled_width],
anchor_boxes: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, 4] representing anchor boxes at each
level.
"""
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
image_height, image_width, _ = image.get_shape().as_list()
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Compute Anchor boxes.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
labels = {
'source_id': dataloader_utils.process_source_id(data['source_id']),
'anchor_boxes': input_anchor.multilevel_boxes,
'image_info': image_info,
}
if self._mode == ModeKeys.PREDICT_WITH_GT:
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(data['groundtruth_boxes'],
image_shape)
groundtruths = {
'source_id': data['source_id'],
'height': data['height'],
'width': data['width'],
'num_detections': tf.shape(data['groundtruth_classes']),
'boxes': boxes,
'classes': data['groundtruth_classes'],
'areas': data['groundtruth_area'],
'is_crowds': tf.cast(data['groundtruth_is_crowd'], tf.int32),
}
groundtruths['source_id'] = dataloader_utils.process_source_id(
groundtruths['source_id'])
groundtruths = dataloader_utils.pad_groundtruths_to_fixed_size(
groundtruths, self._max_num_instances)
labels['groundtruths'] = groundtruths
return {
'images': image,
'labels': labels,
}
def _parse_train_data(self, data):
"""Parse data for ShapeMask training."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
masks = data['groundtruth_instance_masks']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
masks = tf.gather(masks, indices)
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# If not using category, makes all categories with id = 0.
if not self._use_category:
classes = tf.cast(tf.greater(classes, 0), dtype=tf.float32)
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, boxes, masks = input_utils.random_horizontal_flip(
image, boxes, masks)
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
self._output_size,
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
image_scale = image_info[2, :]
offset = image_info[3, :]
# Resizes and crops boxes and masks.
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
self._output_size, offset)
masks = input_utils.resize_and_crop_masks(
tf.expand_dims(masks, axis=-1), image_scale, self._output_size,
offset)
masks = tf.squeeze(masks, axis=-1)
# Filters out ground truth boxes that are all zeros.
indices = input_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
masks = tf.gather(masks, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size, self._output_size)
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# Sample groundtruth masks/boxes/classes for mask branch.
num_masks = tf.shape(masks)[0]
mask_shape = tf.shape(masks)[1:3]
# Pad sampled boxes/masks/classes to a constant batch size.
padded_boxes = input_utils.pad_to_fixed_size(boxes,
self._num_sampled_masks)
padded_classes = input_utils.pad_to_fixed_size(classes,
self._num_sampled_masks)
padded_masks = input_utils.pad_to_fixed_size(masks,
self._num_sampled_masks)
# Randomly sample groundtruth masks for mask branch training. For the image
# without groundtruth masks, it will sample the dummy padded tensors.
rand_indices = tf.random.uniform([self._num_sampled_masks],
minval=0,
maxval=tf.maximum(num_masks, 1),
dtype=tf.dtypes.int32)
sampled_boxes = tf.gather(padded_boxes, rand_indices)
sampled_classes = tf.gather(padded_classes, rand_indices)
sampled_masks = tf.gather(padded_masks, rand_indices)
# Compute mask targets in feature crop. A feature crop fully contains a
# sampled box.
mask_outer_boxes = box_utils.compute_outer_boxes(
sampled_boxes, mask_shape, scale=self._outer_box_scale)
norm_mask_outer_boxes = box_utils.normalize_boxes(
mask_outer_boxes, mask_shape)
# Set sampled_masks shape to [batch_size, height, width, 1].
sampled_masks = tf.expand_dims(sampled_masks, axis=-1)
mask_targets = tf.image.crop_and_resize(
sampled_masks,
norm_mask_outer_boxes,
box_ind=tf.range(self._num_sampled_masks),
crop_size=[self._mask_crop_size, self._mask_crop_size],
method='bilinear',
extrapolation_value=0,
name='train_mask_targets')
mask_targets = tf.where(tf.greater_equal(mask_targets, 0.5),
tf.ones_like(mask_targets),
tf.zeros_like(mask_targets))
mask_targets = tf.squeeze(mask_targets, axis=-1)
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
# For ShapeMask.
'mask_boxes': sampled_boxes,
'mask_outer_boxes': mask_outer_boxes,
'mask_targets': mask_targets,
'mask_classes': sampled_classes,
'mask_is_valid': tf.cast(tf.not_equal(num_masks, 0), tf.int32)
}
return image, labels
def _parse_train_data(self, data):
"""Parses data for training and evaluation."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
image_height, image_width, _ = image.get_shape().as_list()
# Resizes and crops boxes.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
(image_height, image_width),
offset)
# Filters out ground truth boxes that are all zeros.
indices = input_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
}
return image, labels
def _parse_train_data(self, data):
"""Parses data for training.
Args:
data: the decoded tensor dictionary from TfExampleDecoder.
Returns:
image: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
labels: a dictionary of tensors used for training. The following describes
{key: value} pairs in the dictionary.
image_info: a 2D `Tensor` that encodes the information of the image and
the applied preprocessing. It is in the format of
[[original_height, original_width], [scaled_height, scaled_width],
anchor_boxes: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, 4] representing anchor boxes at each level.
rpn_score_targets: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, anchors_per_location]. The height_l and
width_l represent the dimension of class logits at l-th level.
rpn_box_targets: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, anchors_per_location * 4]. The height_l and
width_l represent the dimension of bounding box regression output at
l-th level.
gt_boxes: Groundtruth bounding box annotations. The box is represented
in [y1, x1, y2, x2] format. The coordinates are w.r.t the scaled
image that is fed to the network. The tennsor is padded with -1 to
the fixed dimension [self._max_num_instances, 4].
gt_classes: Groundtruth classes annotations. The tennsor is padded
with -1 to the fixed dimension [self._max_num_instances].
gt_masks: groundtrugh masks cropped by the bounding box and
resized to a fixed size determined by mask_crop_size.
"""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
if self._include_mask:
masks = data['groundtruth_instance_masks']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
tf.greater(tf.size(is_crowds), 0),
lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
lambda: tf.cast(tf.range(num_groundtrtuhs), tf.int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
if self._include_mask:
masks = tf.gather(masks, indices)
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
if self._include_mask:
image, boxes, masks = input_utils.random_horizontal_flip(
image, boxes, masks)
else:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Converts boxes from normalized coordinates to pixel coordinates.
# Now the coordinates of boxes are w.r.t. the original image.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
image_height, image_width, _ = image.get_shape().as_list()
# Resizes and crops boxes.
# Now the coordinates of boxes are w.r.t the scaled image.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
image_info[1, :], offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
if self._include_mask:
masks = tf.gather(masks, indices)
# Transfer boxes to the original image space and do normalization.
cropped_boxes = boxes + tf.tile(tf.expand_dims(offset, axis=0),
[1, 2])
cropped_boxes /= tf.tile(tf.expand_dims(image_scale, axis=0),
[1, 2])
cropped_boxes = box_utils.normalize_boxes(cropped_boxes,
image_shape)
num_masks = tf.shape(masks)[0]
masks = tf.image.crop_and_resize(
tf.expand_dims(masks, axis=-1),
cropped_boxes,
box_indices=tf.range(num_masks, dtype=tf.int32),
crop_size=[self._mask_crop_size, self._mask_crop_size],
method='bilinear')
masks = tf.squeeze(masks, axis=-1)
# Assigns anchor targets.
# Note that after the target assignment, box targets are absolute pixel
# offsets w.r.t. the scaled image.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
anchor_labeler = anchor.RpnAnchorLabeler(input_anchor,
self._rpn_match_threshold,
self._rpn_unmatched_threshold,
self._rpn_batch_size_per_im,
self._rpn_fg_fraction)
rpn_score_targets, rpn_box_targets = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=-1), dtype=tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
'anchor_boxes': input_anchor.multilevel_boxes,
'image_info': image_info,
'rpn_score_targets': rpn_score_targets,
'rpn_box_targets': rpn_box_targets,
}
labels['gt_boxes'] = input_utils.clip_or_pad_to_fixed_size(
boxes, self._max_num_instances, -1)
labels['gt_classes'] = input_utils.clip_or_pad_to_fixed_size(
classes, self._max_num_instances, -1)
if self._include_mask:
labels['gt_masks'] = input_utils.clip_or_pad_to_fixed_size(
masks, self._max_num_instances, -1)
return image, labels
def initiate():
# Load the label map.
print(' - Loading the label map...')
label_map_dict = {}
if 'csv' == 'csv':
with tf.gfile.Open('dataset/fashionpedia_label_map.csv',
'r') as csv_file:
reader = csv.reader(csv_file, delimiter=':')
for row in reader:
if len(row) != 2:
raise ValueError(
'Each row of the csv label map file must be in '
'`id:name` format.')
id_index = int(row[0])
name = row[1]
label_map_dict[id_index] = {
'id': id_index,
'name': name,
}
else:
raise ValueError('Unsupported label map format: {}.'.format('csv'))
params = config_factory.config_generator('attribute_mask_rcnn')
if 'configs/yaml/spinenet49_amrcnn.yaml':
params = params_dict.override_params_dict(
params, 'configs/yaml/spinenet49_amrcnn.yaml', is_strict=True)
params = params_dict.override_params_dict(params, '', is_strict=True)
params.override(
{
'architecture': {
'use_bfloat16': False, # The inference runs on CPU/GPU.
},
},
is_strict=True)
params.validate()
params.lock()
model = model_factory.model_generator(params)
with tf.Graph().as_default():
image_input = tf.placeholder(shape=(), dtype=tf.string)
image = tf.io.decode_image(image_input, channels=3)
image.set_shape([None, None, 3])
image = input_utils.normalize_image(image)
image_size = [640, 640]
image, image_info = input_utils.resize_and_crop_image(
image,
image_size,
image_size,
aug_scale_min=1.0,
aug_scale_max=1.0)
image.set_shape([image_size[0], image_size[1], 3])
# batching.
images = tf.reshape(image, [1, image_size[0], image_size[1], 3])
images_info = tf.expand_dims(image_info, axis=0)
# model inference
outputs = model.build_outputs(images, {'image_info': images_info},
mode=mode_keys.PREDICT)
outputs['detection_boxes'] = (
outputs['detection_boxes'] /
tf.tile(images_info[:, 2:3, :], [1, 1, 2]))
predictions = outputs
# Create a saver in order to load the pre-trained checkpoint.
saver = tf.train.Saver()
sess = tf.Session()
print(' - Loading the checkpoint...')
saver.restore(sess, 'fashionpedia-spinenet-49/model.ckpt')
print(' - Checkpoint Loaded...')
return sess, predictions, image_input
def _parse_eval_data(self, data):
"""Parses data for evaluation.
Args:
data: the decoded tensor dictionary from TfExampleDecoder.
Returns:
image: image tensor that is preproessed to have normalized value and
dimension [output_size[0], output_size[1], 3]
labels: a dictionary of tensors used for training. The following describes
{key: value} pairs in the dictionary.
image_info: a 2D `Tensor` that encodes the information of the image and
the applied preprocessing. It is in the format of
[[original_height, original_width], [scaled_height, scaled_width],
anchor_boxes: ordered dictionary with keys
[min_level, min_level+1, ..., max_level]. The values are tensor with
shape [height_l, width_l, 4] representing anchor boxes at each level.
groundtruths:
source_id: Groundtruth source id.
height: Original image height.
width: Original image width.
boxes: Groundtruth bounding box annotations. The box is represented
in [y1, x1, y2, x2] format. The coordinates are w.r.t the scaled
image that is fed to the network. The tennsor is padded with -1 to
the fixed dimension [self._max_num_instances, 4].
classes: Groundtruth classes annotations. The tennsor is padded
with -1 to the fixed dimension [self._max_num_instances].
areas: Box area or mask area depend on whether mask is present.
is_crowds: Whether the ground truth label is a crowd label.
num_groundtruths: Number of ground truths in the image.
"""
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
image_height, image_width, _ = image.get_shape().as_list()
# Assigns anchor targets.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Sets up groundtruth data for evaluation.
groundtruths = {
'source_id':
data['source_id'],
'height':
data['height'],
'width':
data['width'],
'num_groundtruths':
tf.shape(data['groundtruth_classes']),
'boxes':
box_utils.denormalize_boxes(data['groundtruth_boxes'],
image_shape),
'classes':
data['groundtruth_classes'],
'areas':
data['groundtruth_area'],
'is_crowds':
tf.cast(data['groundtruth_is_crowd'], tf.int32),
}
# TODO(b/143766089): Add ground truth masks for segmentation metrics.
groundtruths['source_id'] = dataloader_utils.process_source_id(
groundtruths['source_id'])
groundtruths = dataloader_utils.pad_groundtruths_to_fixed_size(
groundtruths, self._max_num_instances)
# Packs labels for model_fn outputs.
labels = {
'anchor_boxes': input_anchor.multilevel_boxes,
'image_info': image_info,
'groundtruths': groundtruths,
}
return image, labels
def main(unused_argv):
del unused_argv
# Load the label map.
print(' - Loading the label map...')
label_map_dict = {}
if FLAGS.label_map_format == 'csv':
with tf.gfile.Open(FLAGS.label_map_file, 'r') as csv_file:
reader = csv.reader(csv_file, delimiter=':')
for row in reader:
if len(row) != 2:
raise ValueError(
'Each row of the csv label map file must be in '
'`id:name` format.')
id_index = int(row[0])
name = row[1]
label_map_dict[id_index] = {
'id': id_index,
'name': name,
}
else:
raise ValueError('Unsupported label map format: {}.'.format(
FLAGS.label_mape_format))
params = config_factory.config_generator(FLAGS.model)
if FLAGS.config_file:
params = params_dict.override_params_dict(params,
FLAGS.config_file,
is_strict=True)
params = params_dict.override_params_dict(params,
FLAGS.params_override,
is_strict=True)
params.override(
{
'architecture': {
'use_bfloat16': False, # The inference runs on CPU/GPU.
},
},
is_strict=True)
params.validate()
params.lock()
model = model_factory.model_generator(params)
with tf.Graph().as_default():
image_input = tf.placeholder(shape=(), dtype=tf.string)
image = tf.io.decode_image(image_input, channels=3)
image.set_shape([None, None, 3])
image = input_utils.normalize_image(image)
image_size = [FLAGS.image_size, FLAGS.image_size]
image, image_info = input_utils.resize_and_crop_image(
image,
image_size,
image_size,
aug_scale_min=1.0,
aug_scale_max=1.0)
image.set_shape([image_size[0], image_size[1], 3])
# batching.
images = tf.reshape(image, [1, image_size[0], image_size[1], 3])
images_info = tf.expand_dims(image_info, axis=0)
# model inference
outputs = model.build_outputs(images, {'image_info': images_info},
mode=mode_keys.PREDICT)
# outputs['detection_boxes'] = (
# outputs['detection_boxes'] / tf.tile(images_info[:, 2:3, :], [1, 1, 2]))
predictions = outputs
# Create a saver in order to load the pre-trained checkpoint.
saver = tf.train.Saver()
image_with_detections_list = []
with tf.Session() as sess:
print(' - Loading the checkpoint...')
saver.restore(sess, FLAGS.checkpoint_path)
image_files = tf.gfile.Glob(FLAGS.image_file_pattern)
for i, image_file in enumerate(image_files):
print(' - Processing image %d...' % i)
with tf.gfile.GFile(image_file, 'rb') as f:
image_bytes = f.read()
image = Image.open(image_file)
image = image.convert(
'RGB') # needed for images with 4 channels.
width, height = image.size
np_image = (np.array(image.getdata()).reshape(
height, width, 3).astype(np.uint8))
print(np_image.shape)
predictions_np = sess.run(predictions,
feed_dict={image_input: image_bytes})
logits = predictions_np['logits'][0]
print(logits.shape)
labels = np.argmax(logits.squeeze(), -1)
print(labels.shape)
print(labels)
labels = np.array(Image.fromarray(labels.astype('uint8')))
print(labels.shape)
plt.imshow(labels)
plt.savefig(f"temp-{i}.png")
def main(unused_argv):
del unused_argv
# Load the label map.
print(' - Loading the label map...')
label_map_dict = {}
if FLAGS.label_map_format == 'csv':
with tf.gfile.Open(FLAGS.label_map_file, 'r') as csv_file:
reader = csv.reader(csv_file, delimiter=':')
for row in reader:
if len(row) != 2:
raise ValueError(
'Each row of the csv label map file must be in '
'`id:name` format.')
id_index = int(row[0])
name = row[1]
label_map_dict[id_index] = {
'id': id_index,
'name': name,
}
else:
raise ValueError('Unsupported label map format: {}.'.format(
FLAGS.label_mape_format))
params = config_factory.config_generator(FLAGS.model)
if FLAGS.config_file:
params = params_dict.override_params_dict(params,
FLAGS.config_file,
is_strict=True)
params = params_dict.override_params_dict(params,
FLAGS.params_override,
is_strict=True)
params.validate()
params.lock()
model = model_factory.model_generator(params)
with tf.Graph().as_default():
image_input = tf.placeholder(shape=(), dtype=tf.string)
image = tf.io.decode_image(image_input, channels=3)
image.set_shape([None, None, 3])
image = input_utils.normalize_image(image)
image_size = [FLAGS.image_size, FLAGS.image_size]
image, image_info = input_utils.resize_and_crop_image(
image,
image_size,
image_size,
aug_scale_min=1.0,
aug_scale_max=1.0)
image.set_shape([image_size[0], image_size[1], 3])
# batching.
images = tf.reshape(image, [1, image_size[0], image_size[1], 3])
images_info = tf.expand_dims(image_info, axis=0)
# model inference
outputs = model.build_outputs(images, {'image_info': images_info},
mode=mode_keys.PREDICT)
outputs['detection_boxes'] = (
outputs['detection_boxes'] /
tf.tile(images_info[:, 2:3, :], [1, 1, 2]))
predictions = outputs
# Create a saver in order to load the pre-trained checkpoint.
saver = tf.train.Saver()
image_with_detections_list = []
with tf.Session() as sess:
print(' - Loading the checkpoint...')
saver.restore(sess, FLAGS.checkpoint_path)
image_files = tf.gfile.Glob(FLAGS.image_file_pattern)
for i, image_file in enumerate(image_files):
print(' - Processing image %d...' % i)
with tf.gfile.GFile(image_file, 'rb') as f:
image_bytes = f.read()
image = Image.open(image_file)
image = image.convert(
'RGB') # needed for images with 4 channels.
width, height = image.size
np_image = (np.array(image.getdata()).reshape(
height, width, 3).astype(np.uint8))
predictions_np = sess.run(predictions,
feed_dict={image_input: image_bytes})
num_detections = int(predictions_np['num_detections'][0])
np_boxes = predictions_np['detection_boxes'][
0, :num_detections]
np_scores = predictions_np['detection_scores'][
0, :num_detections]
np_classes = predictions_np['detection_classes'][
0, :num_detections]
np_classes = np_classes.astype(np.int32)
np_masks = None
if 'detection_masks' in predictions_np:
instance_masks = predictions_np['detection_masks'][
0, :num_detections]
np_masks = mask_utils.paste_instance_masks(
instance_masks, box_utils.yxyx_to_xywh(np_boxes),
height, width)
image_with_detections = (
visualization_utils.
visualize_boxes_and_labels_on_image_array(
np_image,
np_boxes,
np_classes,
np_scores,
label_map_dict,
instance_masks=np_masks,
use_normalized_coordinates=False,
max_boxes_to_draw=FLAGS.max_boxes_to_draw,
min_score_thresh=FLAGS.min_score_threshold))
image_with_detections_list.append(image_with_detections)
print(' - Saving the outputs...')
formatted_image_with_detections_list = [
Image.fromarray(image.astype(np.uint8))
for image in image_with_detections_list
]
html_str = '<html>'
image_strs = []
for formatted_image in formatted_image_with_detections_list:
with io.BytesIO() as stream:
formatted_image.save(stream, format='JPEG')
data_uri = base64.b64encode(stream.getvalue()).decode('utf-8')
image_strs.append(
'<img src="data:image/jpeg;base64,{}", height=800>'.format(
data_uri))
images_str = ' '.join(image_strs)
html_str += images_str
html_str += '</html>'
with tf.gfile.GFile(FLAGS.output_html, 'w') as f:
f.write(html_str)
def _parse_eval_data(self, data):
"""Parses data for training and evaluation."""
groundtruths = {}
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(input=image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
image_height, image_width, _ = image.get_shape().as_list()
# Resizes and crops boxes.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
image_info[1, :], offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Sets up groundtruth data for evaluation.
groundtruths = {
'source_id':
data['source_id'],
'num_groundtrtuhs':
tf.shape(data['groundtruth_classes']),
'image_info':
image_info,
'boxes':
box_utils.denormalize_boxes(data['groundtruth_boxes'],
image_shape),
'classes':
data['groundtruth_classes'],
'areas':
data['groundtruth_area'],
'is_crowds':
tf.cast(data['groundtruth_is_crowd'], tf.int32),
}
groundtruths['source_id'] = process_source_id(
groundtruths['source_id'])
groundtruths = pad_groundtruths_to_fixed_size(groundtruths,
self._max_num_instances)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
'groundtruths': groundtruths,
}
return image, labels
def _parse_predict_data(self, data):
"""Parses data for prediction."""
# Gets original image and its size.
image = data['image']
image_shape = tf.shape(image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=1.0,
aug_scale_max=1.0)
image_height, image_width, _ = image.get_shape().as_list()
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Compute Anchor boxes.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
labels = {
'anchor_boxes': input_anchor.multilevel_boxes,
'image_info': image_info,
}
# If mode is PREDICT_WITH_GT, returns groundtruths and training targets
# in labels.
if self._mode == ModeKeys.PREDICT_WITH_GT:
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(data['groundtruth_boxes'],
image_shape)
groundtruths = {
'source_id': data['source_id'],
'height': data['height'],
'width': data['width'],
'num_detections': tf.shape(data['groundtruth_classes']),
'boxes': boxes,
'classes': data['groundtruth_classes'],
'areas': data['groundtruth_area'],
'is_crowds': tf.cast(data['groundtruth_is_crowd'], tf.int32),
}
groundtruths['source_id'] = dataloader_utils.process_source_id(
groundtruths['source_id'])
groundtruths = dataloader_utils.pad_groundtruths_to_fixed_size(
groundtruths, self._max_num_instances)
labels['groundtruths'] = groundtruths
# Computes training objective for evaluation loss.
classes = data['groundtruth_classes']
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(
boxes, image_scale, (image_height, image_width), offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchors.
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
labels['cls_targets'] = cls_targets
labels['box_targets'] = box_targets
labels['num_positives'] = num_positives
return {
'images': image,
'labels': labels,
}
def _parse_train_data(self, data):
"""Parses data for training and evaluation."""
classes = data['groundtruth_classes']
boxes = data['groundtruth_boxes']
is_crowds = data['groundtruth_is_crowd']
# Skips annotations with `is_crowd` = True.
if self._skip_crowd_during_training and self._is_training:
num_groundtrtuhs = tf.shape(input=classes)[0]
with tf.control_dependencies([num_groundtrtuhs, is_crowds]):
indices = tf.cond(
pred=tf.greater(tf.size(input=is_crowds), 0),
true_fn=lambda: tf.where(tf.logical_not(is_crowds))[:, 0],
false_fn=lambda: tf.cast(tf.range(num_groundtrtuhs), tf.
int64))
classes = tf.gather(classes, indices)
boxes = tf.gather(boxes, indices)
# Gets original image and its size.
image = data['image']
# NOTE: The autoaugment method works best when used alongside the standard
# horizontal flipping of images along with size jittering and normalization.
if self._use_autoaugment:
image, boxes = autoaugment_utils.distort_image_with_autoaugment(
image, boxes, self._autoaugment_policy_name)
image_shape = tf.shape(input=image)[0:2]
# Normalizes image with mean and std pixel values.
image = input_utils.normalize_image(image)
# Flips image randomly during training.
if self._aug_rand_hflip:
image, boxes = input_utils.random_horizontal_flip(image, boxes)
# Converts boxes from normalized coordinates to pixel coordinates.
boxes = box_utils.denormalize_boxes(boxes, image_shape)
# Resizes and crops image.
image, image_info = input_utils.resize_and_crop_image(
image,
self._output_size,
padded_size=input_utils.compute_padded_size(
self._output_size, 2**self._max_level),
aug_scale_min=self._aug_scale_min,
aug_scale_max=self._aug_scale_max)
image_height, image_width, _ = image.get_shape().as_list()
# Resizes and crops boxes.
image_scale = image_info[2, :]
offset = image_info[3, :]
boxes = input_utils.resize_and_crop_boxes(boxes, image_scale,
image_info[1, :], offset)
# Filters out ground truth boxes that are all zeros.
indices = box_utils.get_non_empty_box_indices(boxes)
boxes = tf.gather(boxes, indices)
classes = tf.gather(classes, indices)
# Assigns anchors.
input_anchor = anchor.Anchor(self._min_level, self._max_level,
self._num_scales, self._aspect_ratios,
self._anchor_size,
(image_height, image_width))
anchor_labeler = anchor.AnchorLabeler(input_anchor,
self._match_threshold,
self._unmatched_threshold)
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(
boxes, tf.cast(tf.expand_dims(classes, axis=1), tf.float32))
# If bfloat16 is used, casts input image to tf.bfloat16.
if self._use_bfloat16:
image = tf.cast(image, dtype=tf.bfloat16)
# Packs labels for model_fn outputs.
labels = {
'cls_targets': cls_targets,
'box_targets': box_targets,
'anchor_boxes': input_anchor.multilevel_boxes,
'num_positives': num_positives,
'image_info': image_info,
}
# return image, labels
num_anchors = input_anchor.anchors_per_location
mlvl_cls_targets = tf.concat([tf.reshape(cls_targets[lv], [-1, num_anchors]) \
for lv in range(self._min_level, self._max_level+1)], axis=0)
mlvl_box_targets = tf.concat([tf.reshape(box_targets[lv], [-1, num_anchors*4]) \
for lv in range(self._min_level, self._max_level + 1)], axis=0)
num_positives_expand = tf.ones_like(
mlvl_box_targets[..., 0:1]) * num_positives
mlvl_cls_targets_wp = tf.concat(
[mlvl_cls_targets,
tf.cast(num_positives_expand, dtype=tf.int32)],
axis=-1)
mlvl_box_targets_wp = tf.concat(
[mlvl_box_targets, num_positives_expand], axis=-1)
return image, (mlvl_cls_targets_wp, mlvl_box_targets_wp)
