Commit add6e22f authored by Vishnu Banna's avatar Vishnu Banna
Browse files

datapipeline update

parent d09d4bef
# Copyright 2021 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Tensorflow Example proto decoder for object detection.
A decoder to decode string tensors containing serialized tensorflow.Example
protos for object detection.
"""
import tensorflow as tf
from official.vision.beta.dataloaders import tf_example_decoder
def _coco91_to_80(classif, box, areas, iscrowds):
"""Function used to reduce COCO 91 to COCO 80, or to convert from the 2017
foramt to the 2014 format"""
# Vector where index i coralates to the class at index[i].
x = [
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22,
23, 24, 25, 27, 28, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43,
44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62,
63, 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85,
86, 87, 88, 89, 90
]
no = tf.expand_dims(tf.convert_to_tensor(x), axis=0)
# Resahpe the classes to in order to build a class mask.
ce = tf.expand_dims(classif, axis=-1)
# One hot the classificiations to match the 80 class format.
ind = ce == tf.cast(no, ce.dtype)
# Select the max values.
co = tf.reshape(tf.math.argmax(tf.cast(ind, tf.float32), axis=-1), [-1])
ind = tf.where(tf.reduce_any(ind, axis=-1))
# Gather the valuable instances.
classif = tf.gather_nd(co, ind)
box = tf.gather_nd(box, ind)
areas = tf.gather_nd(areas, ind)
iscrowds = tf.gather_nd(iscrowds, ind)
# Restate the number of viable detections, ideally it should be the same.
num_detections = tf.shape(classif)[0]
return classif, box, areas, iscrowds, num_detections
class TfExampleDecoder(tf_example_decoder.TfExampleDecoder):
"""Tensorflow Example proto decoder."""
def __init__(self,
coco91_to_80,
include_mask=False,
regenerate_source_id=False,
mask_binarize_threshold=None):
if coco91_to_80 and include_mask:
raise ValueError("If masks are included you cannot \
convert coco from the 91 class format \
to the 80 class format")
self._coco91_to_80 = coco91_to_80
super().__init__(
include_mask=include_mask,
regenerate_source_id=regenerate_source_id,
mask_binarize_threshold=mask_binarize_threshold
)
def decode(self, serialized_example):
"""Decode the serialized example.
Args:
serialized_example: a single serialized tf.Example string.
Returns:
decoded_tensors: a dictionary of tensors with the following fields:
- source_id: a string scalar tensor.
- image: a uint8 tensor of shape [None, None, 3].
- height: an integer scalar tensor.
- width: an integer scalar tensor.
- groundtruth_classes: a int64 tensor of shape [None].
- groundtruth_is_crowd: a bool tensor of shape [None].
- groundtruth_area: a float32 tensor of shape [None].
- groundtruth_boxes: a float32 tensor of shape [None, 4].
- groundtruth_instance_masks: a float32 tensor of shape
[None, None, None].
- groundtruth_instance_masks_png: a string tensor of shape [None].
"""
decoded_tensors = super().decode(serialized_example)
if self._coco91_to_80:
(decoded_tensors['groundtruth_classes'],
decoded_tensors['groundtruth_boxes'],
decoded_tensors['groundtruth_area'],
decoded_tensors['groundtruth_is_crowd'],
_) = _coco91_to_80(decoded_tensors['groundtruth_classes'],
decoded_tensors['groundtruth_boxes'],
decoded_tensors['groundtruth_area'],
decoded_tensors['groundtruth_is_crowd'])
return decoded_tensors
# Copyright 2021 The TensorFlow Authors. All Rights Reserved. """ Detection Data parser and processing for YOLO.
# Parse image and ground truths in a dataset to training targets and package them
# Licensed under the Apache License, Version 2.0 (the "License"); into (image, labels) tuple for RetinaNet.
# you may not use this file except in compliance with the License. """
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Detection Data parser and processing for YOLO."""
import tensorflow as tf import tensorflow as tf
import numpy as np import numpy as np
from official.vision.beta.projects.yolo.ops import preprocessing_ops from official.vision.beta.projects.yolo.ops import preprocessing_ops
from official.vision.beta.projects.yolo.ops import box_ops as box_utils from official.vision.beta.projects.yolo.ops import anchor
from official.vision.beta.ops import preprocess_ops from official.vision.beta.ops import preprocess_ops
from official.vision.beta.ops import box_ops as bbox_ops
from official.vision.beta.dataloaders import parser, utils from official.vision.beta.dataloaders import parser, utils
def _coco91_to_80(classif, box, areas, iscrowds):
"""Function used to reduce COCO 91 to COCO 80, or to convert from the 2017
foramt to the 2014 format"""
# Vector where index i coralates to the class at index[i].
x = [
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22,
23, 24, 25, 27, 28, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43,
44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62,
63, 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85,
86, 87, 88, 89, 90
]
no = tf.expand_dims(tf.convert_to_tensor(x), axis=0)
# Resahpe the classes to in order to build a class mask.
ce = tf.expand_dims(classif, axis=-1)
# One hot the classificiations to match the 80 class format.
ind = ce == tf.cast(no, ce.dtype)
# Select the max values.
co = tf.reshape(tf.math.argmax(tf.cast(ind, tf.float32), axis=-1), [-1])
ind = tf.where(tf.reduce_any(ind, axis=-1))
# Gather the valuable instances.
classif = tf.gather_nd(co, ind)
box = tf.gather_nd(box, ind)
areas = tf.gather_nd(areas, ind)
iscrowds = tf.gather_nd(iscrowds, ind)
# Restate the number of viable detections, ideally it should be the same.
num_detections = tf.shape(classif)[0]
return classif, box, areas, iscrowds, num_detections
class Parser(parser.Parser): class Parser(parser.Parser):
"""Parse the dataset in to the YOLO model format. """ """Parse the dataset in to the YOLO model format. """
def __init__( def __init__(
self, self,
output_size, output_size,
masks,
anchors, anchors,
strides, expanded_strides,
anchor_free_limits=None, anchor_free_limits=None,
max_num_instances=200, max_num_instances=200,
area_thresh=0.1, area_thresh=0.1,
...@@ -82,23 +37,18 @@ class Parser(parser.Parser): ...@@ -82,23 +37,18 @@ class Parser(parser.Parser):
anchor_t=4.0, anchor_t=4.0,
scale_xy=None, scale_xy=None,
best_match_only=False, best_match_only=False,
coco91to80=False,
darknet=False, darknet=False,
use_tie_breaker=True, use_tie_breaker=True,
dtype='float32', dtype='float32',
seed=None, seed=None):
):
"""Initializes parameters for parsing annotations in the dataset. """Initializes parameters for parsing annotations in the dataset.
Args: Args:
output_size: `Tensor` or `List` for [height, width] of output image. The output_size: `Tensor` or `List` for [height, width] of output image. The
output_size should be divided by the largest feature stride 2^max_level. output_size should be divided by the largest feature stride 2^max_level.
masks: `Dict[List[int]]` of values indicating the indexes in the anchors: `Dict[List[Union[int, float]]]` values for each anchor box.
list of anchor boxes to use an each prediction level between min_level expanded_strides: `Dict[int]` for how much the model scales down the
and max_level. each level must have a list of indexes. images at the largest level.
anchors: `List[List[Union[int, float]]]` values for each anchor box.
strides: `Dict[int]` for how much the model scales down the images at the
largest level.
anchor_free_limits: `List` the box sizes that will be allowed at each FPN anchor_free_limits: `List` the box sizes that will be allowed at each FPN
level as is done in the FCOS and YOLOX paper for anchor free box level as is done in the FCOS and YOLOX paper for anchor free box
assignment. Anchor free will perform worse than Anchor based, but only assignment. Anchor free will perform worse than Anchor based, but only
...@@ -144,9 +94,7 @@ class Parser(parser.Parser): ...@@ -144,9 +94,7 @@ class Parser(parser.Parser):
there should be one value for scale_xy for each level from min_level to there should be one value for scale_xy for each level from min_level to
max_level. max_level.
best_match_only: `boolean` indicating how boxes are selected for best_match_only: `boolean` indicating how boxes are selected for
optimization. optimization.
coco91to80: `bool` for wether to convert coco91 to coco80 to minimize
model parameters.
darknet: `boolean` indicating which data pipeline to use. Setting to True darknet: `boolean` indicating which data pipeline to use. Setting to True
swaps the pipeline to output images realtive to Yolov4 and older. swaps the pipeline to output images realtive to Yolov4 and older.
use_tie_breaker: `boolean` indicating whether to use the anchor threshold use_tie_breaker: `boolean` indicating whether to use the anchor threshold
...@@ -155,25 +103,23 @@ class Parser(parser.Parser): ...@@ -155,25 +103,23 @@ class Parser(parser.Parser):
from {"float32", "float16", "bfloat16"}. from {"float32", "float16", "bfloat16"}.
seed: `int` the seed for random number generation. seed: `int` the seed for random number generation.
""" """
for key in masks.keys(): for key in anchors.keys():
# Assert that the width and height is viable # Assert that the width and height is viable
assert output_size[1] % strides[str(key)] == 0 assert output_size[1] % expanded_strides[str(key)] == 0
assert output_size[0] % strides[str(key)] == 0 assert output_size[0] % expanded_strides[str(key)] == 0
# scale of each FPN level # scale of each FPN level
self._strides = strides self._strides = expanded_strides
# Set the width and height properly and base init: # Set the width and height properly and base init:
self._coco91to80 = coco91to80
self._image_w = output_size[1] self._image_w = output_size[1]
self._image_h = output_size[0] self._image_h = output_size[0]
# Set the anchor boxes and masks for each scale # Set the anchor boxes for each scale
self._anchors = anchors self._anchors = anchors
self._anchor_free_limits = anchor_free_limits self._anchor_free_limits = anchor_free_limits
self._masks = {
key: tf.convert_to_tensor(value) for key, value in masks.items() # anchor labeling paramters
}
self._use_tie_breaker = use_tie_breaker self._use_tie_breaker = use_tie_breaker
self._best_match_only = best_match_only self._best_match_only = best_match_only
self._max_num_instances = max_num_instances self._max_num_instances = max_num_instances
...@@ -202,7 +148,7 @@ class Parser(parser.Parser): ...@@ -202,7 +148,7 @@ class Parser(parser.Parser):
self._darknet = darknet self._darknet = darknet
self._area_thresh = area_thresh self._area_thresh = area_thresh
keys = list(self._masks.keys()) keys = list(self._anchors.keys())
if self._anchor_free_limits is not None: if self._anchor_free_limits is not None:
maxim = 2000 maxim = 2000
...@@ -218,10 +164,15 @@ class Parser(parser.Parser): ...@@ -218,10 +164,15 @@ class Parser(parser.Parser):
# Set the data type based on input string # Set the data type based on input string
self._dtype = dtype self._dtype = dtype
def _get_identity_info(self, image): self._label_builder = anchor.YoloAnchorLabeler(
"""Get an identity image op to pad all info vectors, this is used because anchors = self._anchors,
graph compilation if there are a variable number of info objects in a list. match_threshold=self._anchor_t,
""" best_matches_only=self._best_match_only,
use_tie_breaker=self._use_tie_breaker
)
def _pad_infos_object(self, image):
"""Get a Tensor to pad the info object list."""
shape_ = tf.shape(image) shape_ = tf.shape(image)
val = tf.stack([ val = tf.stack([
tf.cast(shape_[:2], tf.float32), tf.cast(shape_[:2], tf.float32),
...@@ -234,16 +185,16 @@ class Parser(parser.Parser): ...@@ -234,16 +185,16 @@ class Parser(parser.Parser):
def _jitter_scale(self, image, shape, letter_box, jitter, random_pad, def _jitter_scale(self, image, shape, letter_box, jitter, random_pad,
aug_scale_min, aug_scale_max, translate, angle, aug_scale_min, aug_scale_max, translate, angle,
perspective): perspective):
"""Distort and scale each input image"""
infos = []
if (aug_scale_min != 1.0 or aug_scale_max != 1.0): if (aug_scale_min != 1.0 or aug_scale_max != 1.0):
crop_only = True crop_only = True
# jitter gives you only one info object, resize and crop gives you one, # jitter gives you only one info object, resize and crop gives you one,
# if crop only then there can be 1 form jitter and 1 from crop # if crop only then there can be 1 form jitter and 1 from crop
reps = 1 infos.append(self._pad_infos_object(image))
else: else:
crop_only = False crop_only = False
reps = 0 image, crop_info, _ = preprocessing_ops.resize_and_jitter_image(
infos = []
image, info_a, _ = preprocessing_ops.resize_and_jitter_image(
image, image,
shape, shape,
letter_box=letter_box, letter_box=letter_box,
...@@ -252,10 +203,7 @@ class Parser(parser.Parser): ...@@ -252,10 +203,7 @@ class Parser(parser.Parser):
random_pad=random_pad, random_pad=random_pad,
seed=self._seed, seed=self._seed,
) )
infos.extend(info_a) infos.extend(crop_info)
stale_a = self._get_identity_info(image)
for _ in range(reps):
infos.append(stale_a)
image, _, affine = preprocessing_ops.affine_warp_image( image, _, affine = preprocessing_ops.affine_warp_image(
image, image,
shape, shape,
...@@ -269,21 +217,8 @@ class Parser(parser.Parser): ...@@ -269,21 +217,8 @@ class Parser(parser.Parser):
) )
return image, infos, affine return image, infos, affine
def reorg91to80(self, data):
"""Function used to reduce COCO 91 to COCO 80, or to convert from the 2017
foramt to the 2014 format"""
if self._coco91to80:
(data['groundtruth_classes'], data['groundtruth_boxes'],
data['groundtruth_area'], data['groundtruth_is_crowd'],
_) = _coco91_to_80(data['groundtruth_classes'],
data['groundtruth_boxes'], data['groundtruth_area'],
data['groundtruth_is_crowd'])
return data
def _parse_train_data(self, data): def _parse_train_data(self, data):
"""Parses data for training and evaluation.""" """Parses data for training."""
# Down size coco 91 to coco 80 if the option is selected.
data = self.reorg91to80(data)
# Initialize the shape constants. # Initialize the shape constants.
image = data['image'] image = data['image']
...@@ -316,12 +251,16 @@ class Parser(parser.Parser): ...@@ -316,12 +251,16 @@ class Parser(parser.Parser):
else: else:
image = tf.image.resize( image = tf.image.resize(
image, (self._image_h, self._image_w), method='nearest') image, (self._image_h, self._image_w), method='nearest')
inds = tf.cast(tf.range(0, tf.shape(boxes)[0]), tf.int64) output_size = tf.cast([640, 640], tf.float32)
info = self._get_identity_info(image) boxes_ = bbox_ops.denormalize_boxes(boxes, output_size)
inds = bbox_ops.get_non_empty_box_indices(boxes_)
boxes = tf.gather(boxes, inds)
classes = tf.gather(classes, inds)
info = self._pad_infos_object(image)
# Apply scaling to the hue saturation and brightness of an image. # Apply scaling to the hue saturation and brightness of an image.
image = tf.cast(image, dtype=self._dtype) image = tf.cast(image, dtype=self._dtype)
image = image / 255 image = image / 255.0
image = preprocessing_ops.image_rand_hsv( image = preprocessing_ops.image_rand_hsv(
image, image,
self._aug_rand_hue, self._aug_rand_hue,
...@@ -331,30 +270,20 @@ class Parser(parser.Parser): ...@@ -331,30 +270,20 @@ class Parser(parser.Parser):
darknet=self._darknet) darknet=self._darknet)
# Cast the image to the selcted datatype. # Cast the image to the selcted datatype.
image, labels = self._build_label( image, labels = self._build_label(image, boxes, classes,
image, info, inds, data, is_training=True)
boxes,
classes,
self._image_w,
self._image_h,
info,
inds,
data,
is_training=True)
return image, labels return image, labels
def _parse_eval_data(self, data): def _parse_eval_data(self, data):
# Down size coco 91 to coco 80 if the option is selected. """Parses data for evaluation."""
data = self.reorg91to80(data)
# Get the image shape constants and cast the image to the selcted datatype. # Get the image shape constants and cast the image to the selcted datatype.
image = tf.cast(data['image'], dtype=self._dtype) image = tf.cast(data['image'], dtype=self._dtype)
boxes = data['groundtruth_boxes'] boxes = data['groundtruth_boxes']
classes = data['groundtruth_classes'] classes = data['groundtruth_classes']
height, width = self._image_h, self._image_w
image, infos, _ = preprocessing_ops.resize_and_jitter_image( image, infos, _ = preprocessing_ops.resize_and_jitter_image(
image, [height, width], image, [self._image_h, self._image_w],
letter_box=self._letter_box, letter_box=self._letter_box,
random_pad=False, random_pad=False,
shiftx=0.5, shiftx=0.5,
...@@ -362,7 +291,7 @@ class Parser(parser.Parser): ...@@ -362,7 +291,7 @@ class Parser(parser.Parser):
jitter=0.0) jitter=0.0)
# Clip and clean boxes. # Clip and clean boxes.
image = image / 255 image = image / 255.0
boxes, inds = preprocessing_ops.apply_infos( boxes, inds = preprocessing_ops.apply_infos(
boxes, infos, shuffle_boxes=False, area_thresh=0.0, augment=True) boxes, infos, shuffle_boxes=False, area_thresh=0.0, augment=True)
classes = tf.gather(classes, inds) classes = tf.gather(classes, inds)
...@@ -372,8 +301,6 @@ class Parser(parser.Parser): ...@@ -372,8 +301,6 @@ class Parser(parser.Parser):
image, image,
boxes, boxes,
classes, classes,
width,
height,
info, info,
inds, inds,
data, data,
...@@ -381,6 +308,7 @@ class Parser(parser.Parser): ...@@ -381,6 +308,7 @@ class Parser(parser.Parser):
return image, labels return image, labels
def set_shape(self, values, pad_axis=0, pad_value=0, inds=None, scale=1): def set_shape(self, values, pad_axis=0, pad_value=0, inds=None, scale=1):
"""Calls set shape for all input objects."""
if inds is not None: if inds is not None:
values = tf.gather(values, inds) values = tf.gather(values, inds)
vshape = values.get_shape().as_list() vshape = values.get_shape().as_list()
...@@ -396,8 +324,8 @@ class Parser(parser.Parser): ...@@ -396,8 +324,8 @@ class Parser(parser.Parser):
values.set_shape(vshape) values.set_shape(vshape)
return values return values
def _build_grid(self, raw_true, width, height, use_tie_breaker=False): def _build_grid(self, boxes, classes, width, height):
'''Private function for building the full scale object and class grid.''' """Private function for building the full scale object and class grid."""
indexes = {} indexes = {}
updates = {} updates = {}
true_grids = {} true_grids = {}
...@@ -406,27 +334,19 @@ class Parser(parser.Parser): ...@@ -406,27 +334,19 @@ class Parser(parser.Parser):
self._anchor_free_limits = [0.0] + self._anchor_free_limits + [np.inf] self._anchor_free_limits = [0.0] + self._anchor_free_limits + [np.inf]
# for each prediction path generate a properly scaled output prediction map # for each prediction path generate a properly scaled output prediction map
for i, key in enumerate(self._masks.keys()): for i, key in enumerate(self._anchors.keys()):
if self._anchor_free_limits is not None: if self._anchor_free_limits is not None:
fpn_limits = self._anchor_free_limits[i:i + 2] fpn_limits = self._anchor_free_limits[i:i + 2]
else: else:
fpn_limits = None fpn_limits = None
# build the actual grid as well and the list of boxes and classes AND
# their index in the prediction grid
scale_xy = self._scale_xy[key] if not self._darknet else 1 scale_xy = self._scale_xy[key] if not self._darknet else 1
(indexes[key], updates[key],
true_grids[key]) = preprocessing_ops.build_grided_gt_ind( indexes[key], updates[key], true_grids[key] = self._label_builder(
raw_true, key, boxes, classes, self._anchors[key],
self._masks[key], width, height, self._strides[str(key)],
width // self._strides[str(key)], scale_xy, self._max_num_instances * self._scale_up[key],
height // self._strides[str(key)], fpn_limits = fpn_limits)
raw_true['bbox'].dtype,
scale_xy,
self._scale_up[key],
use_tie_breaker,
self._strides[str(key)],
fpn_limits=fpn_limits)
# set/fix the shapes # set/fix the shapes
indexes[key] = self.set_shape(indexes[key], -2, None, None, indexes[key] = self.set_shape(indexes[key], -2, None, None,
...@@ -442,54 +362,39 @@ class Parser(parser.Parser): ...@@ -442,54 +362,39 @@ class Parser(parser.Parser):
image, image,
gt_boxes, gt_boxes,
gt_classes, gt_classes,
width,
height,
info, info,
inds, inds,
data, data,
is_training=True): is_training=True):
"""Label construction for both the train and eval data. """ """Label construction for both the train and eval data. """
width = self._image_w
height = self._image_h
# Set the image shape. # Set the image shape.
imshape = image.get_shape().as_list() imshape = image.get_shape().as_list()
imshape[-1] = 3 imshape[-1] = 3
image.set_shape(imshape) image.set_shape(imshape)
# Get the best anchors. labels = dict()
boxes = box_utils.yxyx_to_xcycwh(gt_boxes) labels['inds'], labels['upds'], labels['true_conf'] = self._build_grid(
best_anchors, ious = preprocessing_ops.get_best_anchor( gt_boxes, gt_classes, width, height)
boxes,
self._anchors,
width=width,
height=height,
iou_thresh=self._anchor_t,
best_match_only=self._best_match_only)
# Set/fix the boxes shape. # Set/fix the boxes shape.
boxes = self.set_shape(boxes, pad_axis=0, pad_value=0) boxes = self.set_shape(gt_boxes, pad_axis=0, pad_value=0)
classes = self.set_shape(gt_classes, pad_axis=0, pad_value=-1) classes = self.set_shape(gt_classes, pad_axis=0, pad_value=-1)
best_anchors = self.set_shape(best_anchors, pad_axis=0, pad_value=-1)
ious = self.set_shape(ious, pad_axis=0, pad_value=0)
area = self.set_shape( area = self.set_shape(
data['groundtruth_area'], pad_axis=0, pad_value=0, inds=inds) data['groundtruth_area'], pad_axis=0, pad_value=0, inds=inds)
is_crowd = self.set_shape( is_crowd = self.set_shape(
data['groundtruth_is_crowd'], pad_axis=0, pad_value=0, inds=inds) data['groundtruth_is_crowd'], pad_axis=0, pad_value=0, inds=inds)
# Build the dictionary set. # Build the dictionary set.
labels = { labels.update({
'source_id': utils.process_source_id(data['source_id']), 'source_id': utils.process_source_id(data['source_id']),
'bbox': tf.cast(boxes, dtype=self._dtype), 'bbox': tf.cast(boxes, dtype=self._dtype),
'classes': tf.cast(classes, dtype=self._dtype), 'classes': tf.cast(classes, dtype=self._dtype),
'best_anchors': tf.cast(best_anchors, dtype=self._dtype), })
'best_iou_match': ious,
}
# Build the grid formatted for loss computation in model output format.
labels['inds'], labels['upds'], labels['true_conf'] = self._build_grid(
labels, width, height, use_tie_breaker=self._use_tie_breaker)
# Update the labels dictionary. # Update the labels dictionary.
labels['bbox'] = box_utils.xcycwh_to_yxyx(labels['bbox'])
if not is_training: if not is_training:
# Sets up groundtruth data for evaluation. # Sets up groundtruth data for evaluation.
groundtruths = { groundtruths = {
...@@ -509,3 +414,5 @@ class Parser(parser.Parser): ...@@ -509,3 +414,5 @@ class Parser(parser.Parser):
groundtruths, self._max_num_instances) groundtruths, self._max_num_instances)
labels['groundtruths'] = groundtruths labels['groundtruths'] = groundtruths
return image, labels return image, labels
import numpy as np
import tensorflow as tf
from tensorflow.python.ops.gen_math_ops import maximum, minimum
from official.vision.beta.projects.yolo.ops import box_ops
from official.vision.beta.projects.yolo.ops import preprocessing_ops
from official.vision.beta.projects.yolo.ops import loss_utils
def get_best_anchor(y_true,
anchors,
stride,
width=1,
height=1,
iou_thresh=0.25,
best_match_only=False,
use_tie_breaker=True):
"""
get the correct anchor that is assoiciated with each box using IOU
Args:
y_true: tf.Tensor[] for the list of bounding boxes in the yolo format
anchors: list or tensor for the anchor boxes to be used in prediction
found via Kmeans
width: int for the image width
height: int for the image height
Return:
tf.Tensor: y_true with the anchor associated with each ground truth
box known
"""
with tf.name_scope('get_best_anchor'):
width = tf.cast(width, dtype=tf.float32)
height = tf.cast(height, dtype=tf.float32)
scaler = tf.convert_to_tensor([width, height])
true_wh = tf.cast(y_true[..., 2:4], dtype=tf.float32) * scaler
anchors = tf.cast(anchors, dtype=tf.float32)/stride
k = tf.shape(anchors)[0]
anchors = tf.concat([tf.zeros_like(anchors), anchors], axis=-1)
truth_comp = tf.concat([tf.zeros_like(true_wh), true_wh], axis=-1)
if iou_thresh >= 1.0:
anchors = tf.expand_dims(anchors, axis=-2)
truth_comp = tf.expand_dims(truth_comp, axis=-3)
aspect = truth_comp[..., 2:4] / anchors[..., 2:4]
aspect = tf.where(tf.math.is_nan(aspect), tf.zeros_like(aspect), aspect)
aspect = tf.maximum(aspect, 1 / aspect)
aspect = tf.where(tf.math.is_nan(aspect), tf.zeros_like(aspect), aspect)
aspect = tf.reduce_max(aspect, axis=-1)
values, indexes = tf.math.top_k(
tf.transpose(-aspect, perm=[1, 0]),
k=tf.cast(k, dtype=tf.int32),
sorted=True)
values = -values
ind_mask = tf.cast(values < iou_thresh, dtype=indexes.dtype)
else:
# iou_raw = box_ops.compute_iou(truth_comp, anchors)
truth_comp = box_ops.xcycwh_to_yxyx(truth_comp)
anchors = box_ops.xcycwh_to_yxyx(anchors)
iou_raw = box_ops.aggregated_comparitive_iou(
truth_comp,
anchors,
iou_type=3,
)
values, indexes = tf.math.top_k(
iou_raw, #tf.transpose(iou_raw, perm=[0, 2, 1]),
k=tf.cast(k, dtype=tf.int32),
sorted=True)
ind_mask = tf.cast(values >= iou_thresh, dtype=indexes.dtype)
# pad the indexs such that all values less than the thresh are -1
# add one, multiply the mask to zeros all the bad locations
# subtract 1 makeing all the bad locations 0.
if best_match_only:
iou_index = ((indexes[..., 0:] + 1) * ind_mask[..., 0:]) - 1
elif use_tie_breaker:
iou_index = tf.concat([
tf.expand_dims(indexes[..., 0], axis=-1),
((indexes[..., 1:] + 1) * ind_mask[..., 1:]) - 1], axis=-1)
else:
iou_index = tf.concat([
tf.expand_dims(indexes[..., 0], axis=-1),
tf.zeros_like(indexes[..., 1:]) - 1], axis=-1)
return tf.cast(iou_index, dtype=tf.float32), tf.cast(values, dtype=tf.float32)
class YoloAnchorLabeler:
def __init__(self,
anchors = None,
match_threshold = 0.25,
best_matches_only = False,
use_tie_breaker = True):
self.anchors = anchors
self.masks = self._get_mask()
self.match_threshold = match_threshold
self.best_matches_only = best_matches_only
self.use_tie_breaker = use_tie_breaker
def _get_mask(self):
masks = {}
start = 0
minimum = int(min(self.anchors.keys()))
maximum = int(max(self.anchors.keys()))
for i in range(minimum, maximum + 1):
per_scale = len(self.anchors[str(i)])
masks[str(i)] = list(range(start, per_scale + start))
start += per_scale
return masks
def _tie_breaking_search(self, anchors, mask, boxes, classes):
mask = tf.cast(tf.reshape(mask, [1, 1, 1, -1]), anchors.dtype)
anchors = tf.expand_dims(anchors, axis=-1)
viable = tf.where(tf.squeeze(anchors == mask, axis = 0))
gather_id, _, anchor_id = tf.split(viable, 3, axis = -1)
boxes = tf.gather_nd(boxes, gather_id)
classes = tf.gather_nd(classes, gather_id)
classes = tf.expand_dims(classes, axis = -1)
classes = tf.cast(classes, boxes.dtype)
anchor_id = tf.cast(anchor_id, boxes.dtype)
return boxes, classes, anchor_id
def _get_anchor_id(self, key, boxes, classes, anchors, width, height, stride):
"""Find the object anchor assignments in an anchor based paradigm. """
# find the best anchor
num_anchors = len(anchors)
if self.best_matches_only:
# get the best anchor for each box
iou_index, _ = get_best_anchor(boxes, anchors, stride,
width=width, height=height,
best_match_only=True,
iou_thresh=self.match_threshold)
mask = range(num_anchors)
else:
# stitch and search boxes across fpn levels
anchorsvec = []
for stitch in self.anchors.keys():
anchorsvec.extend(self.anchors[stitch])
# get the best anchor for each box
iou_index, _ = get_best_anchor(boxes, anchorsvec, stride,
width=width, height=height,
best_match_only=False,
use_tie_breaker=self.use_tie_breaker,
iou_thresh=self.match_threshold)
mask = self.masks[key]
# search for the correct box to use
(boxes,
classes,
anchors) = self._tie_breaking_search(iou_index, mask, boxes, classes)
return boxes, classes, anchors, num_anchors
def _get_centers(self, boxes, classes, anchors, width, height, offset):
"""Find the object center assignments in an anchor based paradigm. """
grid_xy, wh = tf.split(boxes, 2, axis = -1)
wh_scale = tf.cast(tf.convert_to_tensor([width, height]), boxes.dtype)
grid_xy = grid_xy * wh_scale
centers = tf.math.floor(grid_xy)
if offset != 0.0:
clamp = lambda x, ma: tf.maximum(
tf.minimum(x, tf.cast(ma, x.dtype)), tf.zeros_like(x))
grid_xy_index = grid_xy - centers
positive_shift = ((grid_xy_index < offset) & (grid_xy > 1.))
negative_shift = (
(grid_xy_index > (1 - offset)) & (grid_xy < (wh_scale - 1.)))
zero , _ = tf.split(tf.ones_like(positive_shift), 2, axis = -1)
shift_mask = tf.concat(
[zero, positive_shift, negative_shift], axis = -1)
offset = tf.cast([[0, 0], [1, 0],
[0, 1], [-1, 0],
[0, -1]], offset.dtype) * offset
num_shifts = tf.shape(shift_mask)
num_shifts = num_shifts[-1]
boxes = tf.tile(tf.expand_dims(boxes, axis = -2), [1, num_shifts, 1])
classes = tf.tile(tf.expand_dims(classes, axis = -2), [1, num_shifts, 1])
anchors = tf.tile(tf.expand_dims(anchors, axis = -2), [1, num_shifts, 1])
shift_mask = tf.cast(shift_mask, boxes.dtype)
shift_ind = shift_mask * tf.range(0, num_shifts, dtype = boxes.dtype)
shift_ind = shift_ind - (1 - shift_mask)
shift_ind = tf.expand_dims(shift_ind, axis = -1)
boxes_and_centers = tf.concat(
[boxes, classes, anchors, shift_ind], axis = -1)
boxes_and_centers = tf.reshape(boxes_and_centers, [-1, 7])
_, center_ids = tf.split(boxes_and_centers, [6, 1], axis = -1)
#center_ids = tf.squeeze(center_ids, axis = -1)
select = tf.where(center_ids >= 0)
select, _ = tf.split(select, 2, axis = -1)
boxes_and_centers = tf.gather_nd(boxes_and_centers, select)
# center_ids = tf.cast(center_ids, tf.int32)
center_ids = tf.gather_nd(center_ids, select)
center_ids = tf.cast(center_ids, tf.int32)
shifts = tf.gather_nd(offset, center_ids)
boxes, classes, anchors, _ = tf.split(boxes_and_centers,
[4, 1, 1, 1], axis = -1)
grid_xy, _ = tf.split(boxes, 2, axis = -1)
centers = tf.math.floor(grid_xy * wh_scale - shifts)
centers = clamp(centers, wh_scale - 1)
x, y = tf.split(centers, 2, axis = -1)
centers = tf.cast(tf.concat([y, x, anchors], axis = -1), tf.int32)
return boxes, classes, centers
def _get_anchor_free(self,
boxes,
classes,
height,
width,
stride,
fpn_limits,
center_radius=2.5):
"""Find the box assignements in an anchor free paradigm. """
gen = loss_utils.GridGenerator(
masks=None, anchors=[[1, 1]], scale_anchors=stride)
grid_points = gen(width, height, 1, boxes.dtype)[0]
grid_points = tf.squeeze(grid_points, axis=0)
box_list = boxes
class_list = classes
grid_points = (grid_points + 0.5) * stride
x_centers, y_centers = grid_points[..., 0], grid_points[..., 1]
boxes *= (tf.convert_to_tensor([width, height, width, height]) * stride)
tlbr_boxes = box_ops.xcycwh_to_yxyx(boxes)
boxes = tf.reshape(boxes, [1, 1, -1, 4])
tlbr_boxes = tf.reshape(tlbr_boxes, [1, 1, -1, 4])
if self.use_tie_breaker:
area = tf.reduce_prod(boxes[..., 2:], axis = -1)
# check if the box is in the receptive feild of the this fpn level
b_t = y_centers - tlbr_boxes[..., 0]
b_l = x_centers - tlbr_boxes[..., 1]
b_b = tlbr_boxes[..., 2] - y_centers
b_r = tlbr_boxes[..., 3] - x_centers
box_delta = tf.stack([b_t, b_l, b_b, b_r], axis=-1)
if fpn_limits is not None:
max_reg_targets_per_im = tf.reduce_max(box_delta, axis=-1)
gt_min = max_reg_targets_per_im >= fpn_limits[0]
gt_max = max_reg_targets_per_im <= fpn_limits[1]
is_in_boxes = tf.logical_and(gt_min, gt_max)
else:
is_in_boxes = tf.reduce_min(box_delta, axis=-1) > 0.0
is_in_boxes_all = tf.reduce_any(is_in_boxes, axis=(0, 1), keepdims=True)
# check if the center is in the receptive feild of the this fpn level
c_t = y_centers - (boxes[..., 1] - center_radius * stride)
c_l = x_centers - (boxes[..., 0] - center_radius * stride)
c_b = (boxes[..., 1] + center_radius * stride) - y_centers
c_r = (boxes[..., 0] + center_radius * stride) - x_centers
centers_delta = tf.stack([c_t, c_l, c_b, c_r], axis=-1)
is_in_centers = tf.reduce_min(centers_delta, axis=-1) > 0.0
is_in_centers_all = tf.reduce_any(is_in_centers, axis=(0, 1), keepdims=True)
# colate all masks to get the final locations
is_in_index = tf.logical_or(is_in_boxes_all, is_in_centers_all)
is_in_boxes_and_center = tf.logical_and(is_in_boxes, is_in_centers)
is_in_boxes_and_center = tf.logical_and(is_in_index, is_in_boxes_and_center)
if self.use_tie_breaker:
inf = 10000000
boxes_all = tf.cast(is_in_boxes_and_center, area.dtype)
boxes_all = ((boxes_all * area) + ((1 - boxes_all) * inf))
boxes_min = tf.reduce_min(boxes_all, axis = -1, keepdims = True)
boxes_min = tf.where(boxes_min == inf, -1.0, boxes_min)
is_in_boxes_and_center = boxes_all == boxes_min
# construct the index update grid
reps = tf.reduce_sum(tf.cast(is_in_boxes_and_center, tf.int16), axis=-1)
indexes = tf.cast(tf.where(is_in_boxes_and_center), tf.int32)
y, x, t = tf.split(indexes, 3, axis=-1)
boxes = tf.gather_nd(box_list, t)
classes = tf.cast(tf.gather_nd(class_list, t), boxes.dtype)
reps = tf.gather_nd(reps, tf.concat([y, x], axis=-1))
reps = tf.cast(tf.expand_dims(reps, axis=-1), boxes.dtype)
classes = tf.cast(tf.expand_dims(classes, axis=-1), boxes.dtype)
conf = tf.ones_like(classes)
# return the samples and the indexes
samples = tf.concat([boxes, conf, classes], axis=-1)
indexes = tf.concat([y, x, tf.zeros_like(t)], axis=-1)
return indexes, samples
def __call__(self,
key,
boxes,
classes,
anchors,
width,
height,
stride,
scale_xy,
num_instances,
fpn_limits = None):
"""Builds the labels for a single image, not functional in batch mode.
Args:
boxes: `Tensor` of shape [None, 4] indicating the object locations in
an image.
classes: `Tensor` of shape [None] indicating the each objects classes.
anchors: `List[List[int, float]]` representing the anchor boxes to build
the model against.
width: `int` for the images width.
height: `int` for the images height.
stride: `int` for how much the image gets scaled at this level.
scale_xy: `float` for the center shifts to apply when finding center
assignments for a box.
num_instances: `int` for the maximum number of expanded boxes to allow.
fpn_limits: `List[int]` given no anchor boxes this is used to limit the
boxes assied to the each fpn level based on the levels receptive feild.
Returns:
centers: `Tensor` of shape [None, 3] of indexes in the final grid where
boxes are located.
updates: `Tensor` of shape [None, 8] the value to place in the final grid.
full: `Tensor` of [width/stride, height/stride, num_anchors, 1] holding
a mask of where boxes are locates for confidence losses.
"""
boxes = box_ops.yxyx_to_xcycwh(boxes)
width //= stride
height //= stride
width = tf.cast(width, boxes.dtype)
height = tf.cast(height, boxes.dtype)
if fpn_limits is None:
offset = tf.cast(0.5 * (scale_xy - 1), boxes.dtype)
(boxes, classes,
anchors, num_anchors) = self._get_anchor_id(key, boxes, classes, anchors,
width, height, stride)
boxes, classes, centers = self._get_centers(boxes, classes, anchors,
width, height, offset)
ind_mask = tf.ones_like(classes)
updates = tf.concat([boxes, ind_mask, classes], axis = -1)
else:
(centers, updates) = self._get_anchor_free(boxes, classes, height,
width, stride, fpn_limits)
boxes, ind_mask, classes = tf.split(updates, [4, 1, 1], axis = -1)
num_anchors = 1
width = tf.cast(width, tf.int32)
height = tf.cast(height, tf.int32)
full = tf.zeros([height, width, num_anchors, 1], dtype=classes.dtype)
full = tf.tensor_scatter_nd_add(full, centers, ind_mask)
centers = preprocessing_ops.pad_max_instances(
centers, int(num_instances), pad_value=0, pad_axis=0)
updates = preprocessing_ops.pad_max_instances(
updates, int(num_instances), pad_value=0, pad_axis=0)
return centers, updates, full
# Copyright 2021 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Mosaic data aug for YOLO."""
import random import random
import tensorflow as tf import tensorflow as tf
import tensorflow_addons as tfa import tensorflow_addons as tfa
from official.vision.beta.projects.yolo.ops import preprocessing_ops from official.vision.beta.projects.yolo.ops import preprocessing_ops
from official.vision.beta.ops import box_ops from official.vision.beta.ops import box_ops
from official.vision.beta.ops import preprocess_ops
class Mosaic(object): class Mosaic:
"""Stitch together sets of 4 images to generate samples with more boxes.""" """Stitch together sets of 4 images to generate samples with more boxes."""
def __init__(self, def __init__(self,
...@@ -36,6 +23,7 @@ class Mosaic(object): ...@@ -36,6 +23,7 @@ class Mosaic(object):
aug_rand_perspective=0.0, aug_rand_perspective=0.0,
aug_rand_translate=0.0, aug_rand_translate=0.0,
random_pad=False, random_pad=False,
random_flip=False,
area_thresh=0.1, area_thresh=0.1,
seed=None): seed=None):
"""Initializes parameters for mosaic. """Initializes parameters for mosaic.
...@@ -91,6 +79,7 @@ class Mosaic(object): ...@@ -91,6 +79,7 @@ class Mosaic(object):
self._aug_rand_translate = aug_rand_translate self._aug_rand_translate = aug_rand_translate
self._aug_rand_angle = aug_rand_angle self._aug_rand_angle = aug_rand_angle
self._aug_rand_perspective = aug_rand_perspective self._aug_rand_perspective = aug_rand_perspective
self._random_flip = random_flip
self._deterministic = seed != None self._deterministic = seed != None
self._seed = seed if seed is not None else random.randint(0, 2**30) self._seed = seed if seed is not None else random.randint(0, 2**30)
...@@ -116,6 +105,12 @@ class Mosaic(object): ...@@ -116,6 +105,12 @@ class Mosaic(object):
[self._output_size[1] * 2, self._output_size[0] * 2, 3]) [self._output_size[1] * 2, self._output_size[0] * 2, 3])
return cut, ishape return cut, ishape
def _select_ind(self, inds, *args):
items = []
for item in args:
items.append(tf.gather(item, inds))
return items
def _augment_image(self, def _augment_image(self,
image, image,
boxes, boxes,
...@@ -126,13 +121,16 @@ class Mosaic(object): ...@@ -126,13 +121,16 @@ class Mosaic(object):
ys=0.0, ys=0.0,
cut=None): cut=None):
"""Process a single image prior to the application of patching.""" """Process a single image prior to the application of patching."""
# Randomly flip the image horizontally. if self._random_flip:
letter_box = self._letter_box # Randomly flip the image horizontally.
image, boxes, _ = preprocess_ops.random_horizontal_flip(
image, boxes, seed=self._seed)
#augment the image without resizing
image, infos, crop_points = preprocessing_ops.resize_and_jitter_image( image, infos, crop_points = preprocessing_ops.resize_and_jitter_image(
image, [self._output_size[0], self._output_size[1]], image, [self._output_size[0], self._output_size[1]],
random_pad=False, random_pad=False,
letter_box=letter_box, letter_box=self._letter_box,
jitter=self._random_crop, jitter=self._random_crop,
shiftx=xs, shiftx=xs,
shifty=ys, shifty=ys,
...@@ -147,9 +145,7 @@ class Mosaic(object): ...@@ -147,9 +145,7 @@ class Mosaic(object):
shuffle_boxes=False, shuffle_boxes=False,
augment=True, augment=True,
seed=self._seed) seed=self._seed)
classes = tf.gather(classes, inds) classes, is_crowd, area = self._select_ind(inds, classes, is_crowd, area)
is_crowd = tf.gather(is_crowd, inds)
area = tf.gather(area, inds)
return image, boxes, classes, is_crowd, area, crop_points return image, boxes, classes, is_crowd, area, crop_points
def _mosaic_crop_image(self, image, boxes, classes, is_crowd, area): def _mosaic_crop_image(self, image, boxes, classes, is_crowd, area):
...@@ -173,7 +169,11 @@ class Mosaic(object): ...@@ -173,7 +169,11 @@ class Mosaic(object):
boxes = box_ops.denormalize_boxes(boxes, shape[:2]) boxes = box_ops.denormalize_boxes(boxes, shape[:2])
boxes = boxes + tf.cast([ch, cw, ch, cw], boxes.dtype) boxes = boxes + tf.cast([ch, cw, ch, cw], boxes.dtype)
boxes = box_ops.clip_boxes(boxes, shape[:2]) boxes = box_ops.clip_boxes(boxes, shape[:2])
inds = box_ops.get_non_empty_box_indices(boxes)
boxes = box_ops.normalize_boxes(boxes, shape[:2]) boxes = box_ops.normalize_boxes(boxes, shape[:2])
boxes, classes, is_crowd, area = self._select_ind(inds, boxes, classes, is_crowd, area)
# warp and scale the fully stitched sample # warp and scale the fully stitched sample
image, _, affine = preprocessing_ops.affine_warp_image( image, _, affine = preprocessing_ops.affine_warp_image(
...@@ -190,15 +190,9 @@ class Mosaic(object): ...@@ -190,15 +190,9 @@ class Mosaic(object):
# clip and clean boxes # clip and clean boxes
boxes, inds = preprocessing_ops.apply_infos( boxes, inds = preprocessing_ops.apply_infos(
boxes, boxes, None, affine=affine, area_thresh=self._area_thresh,
None,
affine=affine,
area_thresh=self._area_thresh,
augment=True,
seed=self._seed) seed=self._seed)
classes = tf.gather(classes, inds) classes, is_crowd, area = self._select_ind(inds, classes, is_crowd, area)
is_crowd = tf.gather(is_crowd, inds)
area = tf.gather(area, inds)
return image, boxes, classes, is_crowd, area, area return image, boxes, classes, is_crowd, area, area
def scale_boxes(self, patch, ishape, boxes, classes, xs, ys): def scale_boxes(self, patch, ishape, boxes, classes, xs, ys):
...@@ -224,8 +218,6 @@ class Mosaic(object): ...@@ -224,8 +218,6 @@ class Mosaic(object):
sample['image'], sample['groundtruth_boxes'], sample['image'], sample['groundtruth_boxes'],
sample['groundtruth_classes'], sample['groundtruth_is_crowd'], sample['groundtruth_classes'], sample['groundtruth_is_crowd'],
sample['groundtruth_area'], shiftx, shifty, cut) sample['groundtruth_area'], shiftx, shifty, cut)
if cut is None and ishape is None:
cut, ishape = self._generate_cut()
(boxes, classes) = self.scale_boxes(image, ishape, boxes, classes, (boxes, classes) = self.scale_boxes(image, ishape, boxes, classes,
1 - shiftx, 1 - shifty) 1 - shiftx, 1 - shifty)
...@@ -235,7 +227,6 @@ class Mosaic(object): ...@@ -235,7 +227,6 @@ class Mosaic(object):
sample['groundtruth_classes'] = classes sample['groundtruth_classes'] = classes
sample['groundtruth_is_crowd'] = is_crowd sample['groundtruth_is_crowd'] = is_crowd
sample['groundtruth_area'] = area sample['groundtruth_area'] = area
sample['cut'] = cut
sample['shiftx'] = shiftx sample['shiftx'] = shiftx
sample['shifty'] = shifty sample['shifty'] = shifty
sample['crop_points'] = crop_points sample['crop_points'] = crop_points
...@@ -284,7 +275,9 @@ class Mosaic(object): ...@@ -284,7 +275,9 @@ class Mosaic(object):
sample['num_detections'] = tf.shape(sample['groundtruth_boxes'])[1] sample['num_detections'] = tf.shape(sample['groundtruth_boxes'])[1]
sample['is_mosaic'] = tf.cast(1.0, tf.bool) sample['is_mosaic'] = tf.cast(1.0, tf.bool)
del sample['shiftx'], sample['shifty'], sample['crop_points'], sample['cut'] del sample['shiftx']
del sample['shifty']
del sample['crop_points']
return sample return sample
def _mosaic(self, one, two, three, four): def _mosaic(self, one, two, three, four):
...@@ -349,6 +342,7 @@ class Mosaic(object): ...@@ -349,6 +342,7 @@ class Mosaic(object):
def _apply(self, dataset): def _apply(self, dataset):
"""Apply mosaic to an input dataset.""" """Apply mosaic to an input dataset."""
determ = self._deterministic determ = self._deterministic
dataset = dataset.prefetch(tf.data.AUTOTUNE)
one = dataset.shuffle(100, seed=self._seed, reshuffle_each_iteration=True) one = dataset.shuffle(100, seed=self._seed, reshuffle_each_iteration=True)
two = dataset.shuffle( two = dataset.shuffle(
100, seed=self._seed + 1, reshuffle_each_iteration=True) 100, seed=self._seed + 1, reshuffle_each_iteration=True)
......
# Copyright 2021 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Preproceesing operations for YOLO."""
import tensorflow as tf import tensorflow as tf
import numpy as np import numpy as np
import random import random
...@@ -25,18 +11,19 @@ from official.vision.beta.ops import box_ops as bbox_ops ...@@ -25,18 +11,19 @@ from official.vision.beta.ops import box_ops as bbox_ops
PAD_VALUE = 114 PAD_VALUE = 114
GLOBAL_SEED_SET = False GLOBAL_SEED_SET = False
def set_random_seeds(seed=0): def set_random_seeds(seed=0):
"""Sets all accessible global seeds to properly apply randomization. """Sets all accessible global seeds to properly apply randomization.
This is not the same as passing seed as a variable to each call to tf.random. This is not the same as passing the seed as a variable to each call
For more, see the documentation for tf.random on the tensorflow website to tf.random.For more, see the documentation for tf.random on the tensorflow
https://www.tensorflow.org/api_docs/python/tf/random/set_seed. Note that website https://www.tensorflow.org/api_docs/python/tf/random/set_seed. Note
passing seed to each random number generator will not giv you the expected that passing the seed to each random number generator will not give you the
behavior IF you use more than one generator in a single function. expected behavior if you use more than one generator in a single function.
Args: Args:
seed: `Optional[int]` representing the seed you want to use. seed: `Optional[int]` representing the seed you want to use.
""" """
if seed is not None: if seed is not None:
global GLOBAL_SEED_SET global GLOBAL_SEED_SET
os.environ['PYTHONHASHSEED'] = str(seed) os.environ['PYTHONHASHSEED'] = str(seed)
...@@ -47,15 +34,16 @@ def set_random_seeds(seed=0): ...@@ -47,15 +34,16 @@ def set_random_seeds(seed=0):
def get_pad_value(): def get_pad_value():
"""Return the padding value."""
return PAD_VALUE return PAD_VALUE
def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]): def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]):
"""A unified fucntion for consistant random number generation. """A unified function for consistent random number generation.
Equivalent to tf.random.uniform, except that minval and maxval are flipped if Equivalent to tf.random.uniform, except that minval and maxval are flipped if
minval is greater than maxval. Seed Safe random number generator. minval is greater than maxval. Seed Safe random number generator.
Args: Args:
minval: An `int` for a lower or upper endpoint of the interval from which to minval: An `int` for a lower or upper endpoint of the interval from which to
choose the random number. choose the random number.
...@@ -63,8 +51,8 @@ def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]): ...@@ -63,8 +51,8 @@ def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]):
dtype: The output type of the tensor. dtype: The output type of the tensor.
Returns: Returns:
A random tensor of type dtype that falls between minval and maxval excluding A random tensor of type `dtype` that falls between `minval` and `maxval`
the bigger one. excluding the larger one.
""" """
if GLOBAL_SEED_SET: if GLOBAL_SEED_SET:
seed = None seed = None
...@@ -76,18 +64,18 @@ def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]): ...@@ -76,18 +64,18 @@ def rand_uniform_strong(minval, maxval, dtype=tf.float32, seed=None, shape=[]):
def rand_scale(val, dtype=tf.float32, seed=None): def rand_scale(val, dtype=tf.float32, seed=None):
"""Generate a random number for scaling a parameter by multiplication. """Generates a random number for scaling a parameter by multiplication.
Generates a random number for the scale. Half the time, the value is between Generates a random number for the scale. Half of the time, the value is
[1.0, val) with uniformly distributed probability. The other half, the value between [1.0, val) with uniformly distributed probability. In the other half,
is the reciprocal of this value. the value is the reciprocal of this value. The function is identical to the
one in the original implementation:
The function is identical to the one in the original implementation:
https://github.com/AlexeyAB/darknet/blob/a3714d0a/src/utils.c#L708-L713 https://github.com/AlexeyAB/darknet/blob/a3714d0a/src/utils.c#L708-L713
Args: Args:
val: A float representing the maximum scaling allowed. val: A float representing the maximum scaling allowed.
dtype: The output type of the tensor. dtype: The output type of the tensor.
Returns: Returns:
The random scale. The random scale.
""" """
...@@ -99,18 +87,19 @@ def rand_scale(val, dtype=tf.float32, seed=None): ...@@ -99,18 +87,19 @@ def rand_scale(val, dtype=tf.float32, seed=None):
def pad_max_instances(value, instances, pad_value=0, pad_axis=0): def pad_max_instances(value, instances, pad_value=0, pad_axis=0):
"""Pad pr clip the tensor value to a fixed length along a given axis. """Pad or clip the tensor value to a fixed length along a given axis.
Pad a dimension of the tensor to have a maximum number of instances filling Pads a dimension of the tensor to have a maximum number of instances filling
additional entries with the `pad_value`. Allows for selection of the padding additional entries with the `pad_value`. Allows for selection of the padding
axis axis.
Args: Args:
value: An input tensor. value: An input tensor.
instances: An int representing the maximum number of instances. instances: An `int` representing the maximum number of instances.
pad_value: An int representing the value used for padding until the maximum pad_value: An `int` representing the value used for padding until the
number of instances is obtained. maximum number of instances is obtained.
pad_axis: An int representing the axis index to pad. pad_axis: An `int` representing the axis index to pad.
Returns: Returns:
The output tensor whose dimensions match the input tensor except with the The output tensor whose dimensions match the input tensor except with the
size along the `pad_axis` replaced by `instances`. size along the `pad_axis` replaced by `instances`.
...@@ -137,16 +126,17 @@ def pad_max_instances(value, instances, pad_value=0, pad_axis=0): ...@@ -137,16 +126,17 @@ def pad_max_instances(value, instances, pad_value=0, pad_axis=0):
def get_image_shape(image): def get_image_shape(image):
""" Consitently get the width and height of the image. """Consistently gets the width and height of the image.
Get the shape of the image regardless of if the image is in the Gets the shape of the image regardless of if the image is in the
(batch_size, x, y, c) format or the (x, y, c) format. (batch_size, x, y, c) format or the (x, y, c) format.
Args: Args:
image: A tensor who has either 3 or 4 dimensions. image: A tensor who has either 3 or 4 dimensions.
Returns: Returns:
A tuple representing the (height, width) of the image. A tuple (height, width), where height is the height of the image
and width is the width of the image.
""" """
shape = tf.shape(image) shape = tf.shape(image)
if shape.get_shape().as_list()[0] == 4: if shape.get_shape().as_list()[0] == 4:
...@@ -159,23 +149,7 @@ def get_image_shape(image): ...@@ -159,23 +149,7 @@ def get_image_shape(image):
def _augment_hsv_darknet(image, rh, rs, rv, seed=None): def _augment_hsv_darknet(image, rh, rs, rv, seed=None):
"""Randomly alter the hue, saturation, and brightness of an image. """Randomize the hue, saturation, and brightness via the darknet method."""
Applies ranomdization the same way as Darknet by scaling the saturation and
brightness of the image and adding/rotating the hue.
Args:
image: Tensor of shape [None, None, 3] that needs to be altered.
rh: `float32` used to indicate the maximum delta that can be added to hue.
rs: `float32` used to indicate the maximum delta that can be multiplied to
saturation.
rv: `float32` used to indicate the maximum delta that can be multiplied to
brightness.
seed: `Optional[int]` for the seed to use in random number generation.
Returns:
The HSV altered image in the same datatype as the input image
"""
if rh > 0.0: if rh > 0.0:
delta = rand_uniform_strong(-rh, rh, seed=seed) delta = rand_uniform_strong(-rh, rh, seed=seed)
image = tf.image.adjust_hue(image, delta) image = tf.image.adjust_hue(image, delta)
...@@ -192,24 +166,7 @@ def _augment_hsv_darknet(image, rh, rs, rv, seed=None): ...@@ -192,24 +166,7 @@ def _augment_hsv_darknet(image, rh, rs, rv, seed=None):
def _augment_hsv_torch(image, rh, rs, rv, seed=None): def _augment_hsv_torch(image, rh, rs, rv, seed=None):
"""Randomly alter the hue, saturation, and brightness of an image. """Randomize the hue, saturation, and brightness via the pytorch method."""
Applies ranomdization the same way as Darknet by scaling the saturation and
brightness and hue of the image.
Args:
image: Tensor of shape [None, None, 3] that needs to be altered.
rh: `float32` used to indicate the maximum delta that can be multiplied to
hue.
rs: `float32` used to indicate the maximum delta that can be multiplied to
saturation.
rv: `float32` used to indicate the maximum delta that can be multiplied to
brightness.
seed: `Optional[int]` for the seed to use in random number generation.
Returns:
The HSV altered image in the same datatype as the input image
"""
dtype = image.dtype dtype = image.dtype
image = tf.cast(image, tf.float32) image = tf.cast(image, tf.float32)
image = tf.image.rgb_to_hsv(image) image = tf.image.rgb_to_hsv(image)
...@@ -218,7 +175,6 @@ def _augment_hsv_torch(image, rh, rs, rv, seed=None): ...@@ -218,7 +175,6 @@ def _augment_hsv_torch(image, rh, rs, rv, seed=None):
r = rand_uniform_strong( r = rand_uniform_strong(
-1, 1, shape=[3], dtype=image.dtype, seed=seed) * gen_range + 1 -1, 1, shape=[3], dtype=image.dtype, seed=seed) * gen_range + 1
# image = tf.cast(tf.cast(image, r.dtype) * (r * scale), tf.int32)
image = tf.math.floor(tf.cast(image, scale.dtype) * scale) image = tf.math.floor(tf.cast(image, scale.dtype) * scale)
image = tf.math.floor(tf.cast(image, r.dtype) * r) image = tf.math.floor(tf.cast(image, r.dtype) * r)
h, s, v = tf.split(image, 3, axis=-1) h, s, v = tf.split(image, 3, axis=-1)
...@@ -233,23 +189,24 @@ def _augment_hsv_torch(image, rh, rs, rv, seed=None): ...@@ -233,23 +189,24 @@ def _augment_hsv_torch(image, rh, rs, rv, seed=None):
def image_rand_hsv(image, rh, rs, rv, seed=None, darknet=False): def image_rand_hsv(image, rh, rs, rv, seed=None, darknet=False):
"""Randomly alter the hue, saturation, and brightness of an image. """Randomly alters the hue, saturation, and brightness of an image.
Args: Args:
image: Tensor of shape [None, None, 3] that needs to be altered. image: `Tensor` of shape [None, None, 3] that needs to be altered.
rh: `float32` used to indicate the maximum delta that can be multiplied to rh: `float32` used to indicate the maximum delta that can be multiplied to
hue. the hue.
rs: `float32` used to indicate the maximum delta that can be multiplied to rs: `float32` used to indicate the maximum delta that can be multiplied to
saturation. the saturation.
rv: `float32` used to indicate the maximum delta that can be multiplied to rv: `float32` used to indicate the maximum delta that can be multiplied to
brightness. the brightness.
seed: `Optional[int]` for the seed to use in random number generation. seed: `Optional[int]` for the seed to use in the random number generation.
darknet: `bool` indicating wether the model was orignally built in the darknet: `bool` indicating whether the model was originally built in the
darknet or the pytorch library. Darknet or PyTorch library.
Returns: Returns:
The HSV altered image in the same datatype as the input image The HSV altered image in the same datatype as the input image.
""" """
if darknet: if darknet:
image = _augment_hsv_darknet(image, rh, rs, rv, seed=seed) image = _augment_hsv_darknet(image, rh, rs, rv, seed=seed)
else: else:
...@@ -259,27 +216,27 @@ def image_rand_hsv(image, rh, rs, rv, seed=None, darknet=False): ...@@ -259,27 +216,27 @@ def image_rand_hsv(image, rh, rs, rv, seed=None, darknet=False):
def mosaic_cut(image, original_width, original_height, width, height, center, def mosaic_cut(image, original_width, original_height, width, height, center,
ptop, pleft, pbottom, pright, shiftx, shifty): ptop, pleft, pbottom, pright, shiftx, shifty):
"""Use a provided center to take slices of 4 images to apply mosaic. """Generates a random center location to use for the mosaic operation.
Given a center location, cut the input image into a slice that will be Given a center location, cuts the input image into a slice that will be
concatnated with other slices with the same center in order to construct concatenated with other slices with the same center in order to construct
a final mosaiced image. a final mosaicked image.
Args: Args:
image: Tensor of shape [None, None, 3] that needs to be altered. image: `Tensor` of shape [None, None, 3] that needs to be altered.
original_width: `float` value indicating the orignal width of the image. ow: `float` value indicating the original width of the image.
original_height: `float` value indicating the orignal height of the image. oh: `float` value indicating the original height of the image.
width: `float` value indicating the final width image. w: `float` value indicating the final width of the image.
height: `float` value indicating the final height image. h: `float` value indicating the final height of the image.
center: `float` value indicating the desired center of the final patched center: `float` value indicating the desired center of the final patched
image. image.
ptop: `float` value indicating the top of the image without padding. ptop: `float` value indicating the top of the image without padding.
pleft: `float` value indicating the left of the image without padding. pleft: `float` value indicating the left of the image without padding.
pbottom: `float` value indicating the bottom of the image without padding. pbottom: `float` value indicating the bottom of the image without padding.
pright: `float` value indicating the right of the image without padding. pright: `float` value indicating the right of the image without padding.
shiftx: `float` 0.0 or 1.0 value indicating if the image is in the shiftx: `float` 0.0 or 1.0 value indicating if the image is on the
left or right. left or right.
shifty: `float` 0.0 or 1.0 value indicating if the image is in the shifty: `float` 0.0 or 1.0 value indicating if the image is at the
top or bottom. top or bottom.
Returns: Returns:
...@@ -362,6 +319,39 @@ def resize_and_jitter_image(image, ...@@ -362,6 +319,39 @@ def resize_and_jitter_image(image,
seed=None): seed=None):
"""Resize, Pad, and distort a given input image following Darknet. """Resize, Pad, and distort a given input image following Darknet.
Resizes the input image to output size (RetinaNet style).
Resize and pad images given the desired output size of the image and
stride size.
Here are the preprocessing steps.
1. For a given image, keep its aspect ratio and rescale the image to make it
the largest rectangle to be bounded by the rectangle specified by the
`desired_size`.
2. Pad the rescaled image to the padded_size.
Args:
image: a `Tensor` of shape [height, width, 3] representing an image.
desired_size: a `Tensor` or `int` list/tuple of two elements representing
[height, width] of the desired actual output image size.
padded_size: a `Tensor` or `int` list/tuple of two elements representing
[height, width] of the padded output image size. Padding will be applied
after scaling the image to the desired_size.
aug_scale_min: a `float` with range between [0, 1.0] representing minimum
random scale applied to desired_size for training scale jittering.
aug_scale_max: a `float` with range between [1.0, inf] representing maximum
random scale applied to desired_size for training scale jittering.
seed: seed for random scale jittering.
method: function to resize input image to scaled image.
Returns:
output_image: `Tensor` of shape [height, width, 3] where [height, width]
equals to `output_size`.
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], [desired_height, desired_width],
[y_scale, x_scale], [y_offset, x_offset]], where [desired_height,
desired_width] is the actual scaled image size, and [y_scale, x_scale] is
the scaling factor, which is the ratio of
scaled dimension / original dimension.
""" """
def intersection(a, b): def intersection(a, b):
...@@ -525,7 +515,7 @@ def _build_transform(image, ...@@ -525,7 +515,7 @@ def _build_transform(image,
random_pad=False, random_pad=False,
desired_size=None, desired_size=None,
seed=None): seed=None):
"""Builds a unifed affine transformation to spatially augment the image.""" """Builds a unified affine transformation to spatially augment the image."""
height, width = get_image_shape(image) height, width = get_image_shape(image)
ch = height = tf.cast(height, tf.float32) ch = height = tf.cast(height, tf.float32)
...@@ -624,6 +614,30 @@ def affine_warp_image(image, ...@@ -624,6 +614,30 @@ def affine_warp_image(image,
translate=0.0, translate=0.0,
random_pad=False, random_pad=False,
seed=None): seed=None):
"""Applies random spatial augmentation to the image.
Args:
image: A `Tensor` for the image.
desired_size: A `tuple` for desired output image size.
perspective: An `int` for the maximum that can be applied to random
perspective change.
degrees: An `int` for the maximum degrees that can be applied to random
rotation.
scale_min: An `int` for the minimum scaling factor that can be
applied to random scaling.
scale_max: An `int` for the maximum scaling factor that can be
applied to random scaling.
translate: An `int` for the maximum translation that can be applied to
random translation.
random_pad: A `bool` for using random padding.
seed: An `Optional[int]` for the seed to use in random number generation.
Returns:
image: A `Tensor` representing the augmented image.
affine_matrix: A `Tensor` representing the augmenting matrix for the image.
affine_info: A `List` containing the size of the original image, the desired
output_size of the image and the augmenting matrix for the boxes.
"""
# Build an image transformation matrix. # Build an image transformation matrix.
image_size = tf.cast(get_image_shape(image), tf.float32) image_size = tf.cast(get_image_shape(image), tf.float32)
...@@ -649,11 +663,30 @@ def affine_warp_image(image, ...@@ -649,11 +663,30 @@ def affine_warp_image(image,
interpolation='bilinear') interpolation='bilinear')
desired_size = tf.cast(desired_size, tf.float32) desired_size = tf.cast(desired_size, tf.float32)
return image, affine_matrix, [image_size, desired_size, affine_boxes] affine_info = [image_size, desired_size, affine_boxes]
return image, affine_matrix, affine_info
# ops for box clipping and cleaning # ops for box clipping and cleaning
def affine_warp_boxes(affine, boxes, output_size, box_history): def affine_warp_boxes(affine, boxes, output_size, box_history):
"""Applies random rotation, random perspective change and random translation
and random scaling to the boxes.
Args:
Mb: A `Tensor` for the augmenting matrix for the boxes.
boxes: A `Tensor` for the boxes.
output_size: A `list` of two integers, a two-element vector or a tensor
such that all but the last dimensions are `broadcastable` to `boxes`.
The last dimension is 2, which represents [height, width].
box_history: A `Tensor` for the boxes history, which are the boxes that
undergo the same augmentations as `boxes`, but no clipping was applied.
We can keep track of how much changes are done to the boxes by keeping
track of this tensor.
Returns:
clipped_boxes: A `Tensor` representing the augmented boxes.
box_history: A `Tensor` representing the augmented box_history.
"""
def _get_corners(box): def _get_corners(box):
"""Get the corner of each box as a tuple of (x, y) coordinates""" """Get the corner of each box as a tuple of (x, y) coordinates"""
...@@ -705,6 +738,22 @@ def boxes_candidates(clipped_boxes, ...@@ -705,6 +738,22 @@ def boxes_candidates(clipped_boxes,
wh_thr=2, wh_thr=2,
ar_thr=20, ar_thr=20,
area_thr=0.1): area_thr=0.1):
"""Filters the boxes and keeps the boxes that satisfy thewidth/height and
area constraints.
Args:
clipped_boxes: A `Tensor` for the boxes.
box_history: A `Tensor` for the boxes history, which are the boxes that
undergo the same augmentations as `boxes`, but no clipping was applied.
We can keep track of how much changes are done to the boxes by keeping
track of this tensor.
wh_thr: An `int` for the width/height threshold.
ar_thr: An `int` for the aspect ratio threshold.
area_thr: An `int` for the area threshold.
Returns:
indices[:, 0]: A `Tensor` representing valid boxes after filtering.
"""
area_thr = tf.math.abs(area_thr) area_thr = tf.math.abs(area_thr)
...@@ -743,6 +792,25 @@ def boxes_candidates(clipped_boxes, ...@@ -743,6 +792,25 @@ def boxes_candidates(clipped_boxes,
def resize_and_crop_boxes(boxes, image_scale, output_size, offset, box_history): def resize_and_crop_boxes(boxes, image_scale, output_size, offset, box_history):
"""Resizes and crops the boxes.
Args:
boxes: A `Tensor` for the boxes.
image_scale: A `Tensor` for the scaling factor of the image.
output_size: A `list` of two integers, a two-element vector or a tensor such
that all but the last dimensions are `broadcastable` to `boxes`. The last
dimension is 2, which represents [height, width].
offset: A `Tensor` for how much translation was applied to the image.
box_history: A `Tensor` for the boxes history, which are the boxes that
undergo the same augmentations as `boxes`, but no clipping was applied.
We can keep track of how much changes are done to the boxes by keeping
track of this tensor.
Returns:
clipped_boxes: A `Tensor` representing the augmented boxes.
box_history: A `Tensor` representing the augmented box_history.
"""
# Shift and scale the input boxes. # Shift and scale the input boxes.
boxes *= tf.tile(tf.expand_dims(image_scale, axis=0), [1, 2]) boxes *= tf.tile(tf.expand_dims(image_scale, axis=0), [1, 2])
boxes -= tf.tile(tf.expand_dims(offset, axis=0), [1, 2]) boxes -= tf.tile(tf.expand_dims(offset, axis=0), [1, 2])
...@@ -763,6 +831,22 @@ def apply_infos(boxes, ...@@ -763,6 +831,22 @@ def apply_infos(boxes,
area_thresh=0.1, area_thresh=0.1,
seed=None, seed=None,
augment=True): augment=True):
"""Clips and cleans the boxes.
Args:
boxes: A `Tensor` for the boxes.
image_scale: A `list` that contains the information of the image.
affine: A `list` that contains parameters for resize and crop.
shuffle_boxes: A `bool` for shuffling the boxes.
area_thresh: An `int` for the area threshold.
seed: seed for random number generation.
augment: A `bool` for clipping the boxes to [0, 1].
Returns:
boxes: A `Tensor` representing the augmented boxes.
ind: A `Tensor` valid box indices.
"""
# Clip and clean boxes. # Clip and clean boxes.
def get_valid_boxes(boxes): def get_valid_boxes(boxes):
"""Get indices for non-empty boxes.""" """Get indices for non-empty boxes."""
...@@ -842,388 +926,4 @@ def apply_infos(boxes, ...@@ -842,388 +926,4 @@ def apply_infos(boxes,
boxes_ = bbox_ops.denormalize_boxes(boxes, output_size) boxes_ = bbox_ops.denormalize_boxes(boxes, output_size)
inds = bbox_ops.get_non_empty_box_indices(boxes_) inds = bbox_ops.get_non_empty_box_indices(boxes_)
boxes = tf.gather(boxes, inds) boxes = tf.gather(boxes, inds)
return boxes, inds return boxes, inds
\ No newline at end of file
def _gen_viable_box_mask(boxes):
"""Generate a mask to filter the boxes to only those with in the image. """
equal = tf.reduce_all(tf.math.less_equal(boxes[..., 2:4], 0), axis=-1)
lower_bound = tf.reduce_any(tf.math.less(boxes[..., 0:2], 0.0), axis=-1)
upper_bound = tf.reduce_any(
tf.math.greater_equal(boxes[..., 0:2], 1.0), axis=-1)
negative_mask = tf.logical_or(tf.logical_or(equal, lower_bound), upper_bound)
return tf.logical_not(negative_mask)
def _get_box_locations(anchors, mask, boxes):
"""Calculate the number of anchors associated with each ground truth box."""
box_mask = _gen_viable_box_mask(boxes)
mask = tf.reshape(mask, [1, 1, 1, -1])
box_mask = tf.reshape(box_mask, [-1, 1, 1])
anchors = tf.expand_dims(anchors, axis=-1)
# split the anchors into the best matches and other wise
anchors_primary, anchors_alternate = tf.split(anchors, [1, -1], axis=-2)
anchors_alternate = tf.concat(
[-tf.ones_like(anchors_primary), anchors_alternate], axis=-2)
# convert all the masks into index locations
viable_primary = tf.where(
tf.squeeze(tf.logical_and(box_mask, anchors_primary == mask), axis=0))
viable_alternate = tf.where(
tf.squeeze(tf.logical_and(box_mask, anchors_alternate == mask), axis=0))
viable_full = tf.where(
tf.squeeze(tf.logical_and(box_mask, anchors == mask), axis=0))
# compute the number of anchors associated with each ground truth box.
acheck = tf.reduce_any(anchors == mask, axis=-1)
repititions = tf.squeeze(
tf.reduce_sum(tf.cast(acheck, mask.dtype), axis=-1), axis=0)
# cast to int32
viable_primary = tf.cast(viable_primary, tf.int32)
viable_alternate = tf.cast(viable_alternate, tf.int32)
viable_full = tf.cast(viable_full, tf.int32)
return repititions, viable_primary, viable_alternate, viable_full
def _write_sample(box, anchor_id, offset, sample, ind_val, ind_sample, height,
width, num_written):
"""Find the correct x,y indexs for each box in the output groundtruth."""
anchor_index = tf.convert_to_tensor([tf.cast(anchor_id, tf.int32)])
gain = tf.cast(tf.convert_to_tensor([width, height]), box.dtype)
y = box[1] * height
x = box[0] * width
y_index = tf.convert_to_tensor([tf.cast(y, tf.int32)])
x_index = tf.convert_to_tensor([tf.cast(x, tf.int32)])
grid_idx = tf.concat([y_index, x_index, anchor_index], axis=-1)
ind_val = ind_val.write(num_written, grid_idx)
ind_sample = ind_sample.write(num_written, sample)
num_written += 1
if offset > 0:
offset = tf.cast(offset, x.dtype)
grid_xy = tf.cast(tf.convert_to_tensor([x, y]), x.dtype)
clamp = lambda x, ma: tf.maximum(
tf.minimum(x, tf.cast(ma, x.dtype)), tf.zeros_like(x))
grid_xy_index = grid_xy - tf.floor(grid_xy)
positive_shift = ((grid_xy_index < offset) & (grid_xy > 1.))
negative_shift = ((grid_xy_index > (1 - offset)) & (grid_xy < (gain - 1.)))
shifts = [
positive_shift[0], positive_shift[1], negative_shift[0],
negative_shift[1]
]
offset = tf.cast([[1, 0], [0, 1], [-1, 0], [0, -1]], offset.dtype) * offset
for i in range(4):
if shifts[i]:
x_index = tf.convert_to_tensor([tf.cast(x - offset[i, 0], tf.int32)])
y_index = tf.convert_to_tensor([tf.cast(y - offset[i, 1], tf.int32)])
grid_idx = tf.concat([
clamp(y_index, height - 1),
clamp(x_index, width - 1), anchor_index
],
axis=-1)
ind_val = ind_val.write(num_written, grid_idx)
ind_sample = ind_sample.write(num_written, sample)
num_written += 1
return ind_val, ind_sample, num_written
def _write_grid(viable, num_reps, boxes, classes, ious, ind_val, ind_sample,
height, width, num_written, num_instances, offset):
"""Iterate all viable anchor boxes and write each sample to groundtruth."""
const = tf.cast(tf.convert_to_tensor([1.]), dtype=boxes.dtype)
num_viable = tf.shape(viable)[0]
for val in range(num_viable):
idx = viable[val]
obj_id, anchor, anchor_idx = idx[0], idx[1], idx[2]
if num_written >= num_instances:
break
reps = tf.convert_to_tensor([num_reps[obj_id]])
box = boxes[obj_id]
cls_ = classes[obj_id]
iou = tf.convert_to_tensor([ious[obj_id, anchor]])
sample = tf.concat([box, const, cls_, iou, reps], axis=-1)
ind_val, ind_sample, num_written = _write_sample(box, anchor_idx, offset,
sample, ind_val,
ind_sample, height, width,
num_written)
return ind_val, ind_sample, num_written
def _write_anchor_free_grid(boxes,
classes,
height,
width,
num_written,
stride,
fpn_limits,
center_radius=2.5):
"""Iterate all boxes and write to grid without anchors boxes."""
gen = loss_utils.GridGenerator(
masks=None, anchors=[[1, 1]], scale_anchors=stride)
grid_points = gen(width, height, 1, boxes.dtype)[0]
grid_points = tf.squeeze(grid_points, axis=0)
box_list = boxes
class_list = classes
grid_points = (grid_points + 0.5) * stride
x_centers, y_centers = grid_points[..., 0], grid_points[..., 1]
boxes *= (tf.convert_to_tensor([width, height, width, height]) * stride)
tlbr_boxes = box_ops.xcycwh_to_yxyx(boxes)
boxes = tf.reshape(boxes, [1, 1, -1, 4])
tlbr_boxes = tf.reshape(tlbr_boxes, [1, 1, -1, 4])
mask = tf.reshape(class_list != -1, [1, 1, -1])
# check if the box is in the receptive feild of the this fpn level
b_t = y_centers - tlbr_boxes[..., 0]
b_l = x_centers - tlbr_boxes[..., 1]
b_b = tlbr_boxes[..., 2] - y_centers
b_r = tlbr_boxes[..., 3] - x_centers
box_delta = tf.stack([b_t, b_l, b_b, b_r], axis=-1)
if fpn_limits is not None:
max_reg_targets_per_im = tf.reduce_max(box_delta, axis=-1)
gt_min = max_reg_targets_per_im >= fpn_limits[0]
gt_max = max_reg_targets_per_im <= fpn_limits[1]
is_in_boxes = tf.logical_and(gt_min, gt_max)
else:
is_in_boxes = tf.reduce_min(box_delta, axis=-1) > 0.0
is_in_boxes = tf.logical_and(is_in_boxes, mask)
is_in_boxes_all = tf.reduce_any(is_in_boxes, axis=(0, 1), keepdims=True)
# check if the center is in the receptive feild of the this fpn level
c_t = y_centers - (boxes[..., 1] - center_radius * stride)
c_l = x_centers - (boxes[..., 0] - center_radius * stride)
c_b = (boxes[..., 1] + center_radius * stride) - y_centers
c_r = (boxes[..., 0] + center_radius * stride) - x_centers
centers_delta = tf.stack([c_t, c_l, c_b, c_r], axis=-1)
is_in_centers = tf.reduce_min(centers_delta, axis=-1) > 0.0
is_in_centers = tf.logical_and(is_in_centers, mask)
is_in_centers_all = tf.reduce_any(is_in_centers, axis=(0, 1), keepdims=True)
# colate all masks to get the final locations
is_in_index = tf.logical_or(is_in_boxes_all, is_in_centers_all)
is_in_boxes_and_center = tf.logical_and(is_in_boxes, is_in_centers)
is_in_boxes_and_center = tf.logical_and(is_in_index, is_in_boxes_and_center)
# construct the index update grid
reps = tf.reduce_sum(tf.cast(is_in_boxes_and_center, tf.int16), axis=-1)
indexes = tf.cast(tf.where(is_in_boxes_and_center), tf.int32)
y, x, t = tf.split(indexes, 3, axis=-1)
boxes = tf.gather_nd(box_list, t)
classes = tf.cast(tf.gather_nd(class_list, t), boxes.dtype)
reps = tf.gather_nd(reps, tf.concat([y, x], axis=-1))
reps = tf.cast(tf.expand_dims(reps, axis=-1), boxes.dtype)
conf = tf.ones_like(classes)
# return the samples and the indexes
samples = tf.concat([boxes, conf, classes, conf, reps], axis=-1)
indexes = tf.concat([y, x, tf.zeros_like(t)], axis=-1)
num_written = tf.shape(reps)[0]
return indexes, samples, num_written
def build_grided_gt_ind(y_true,
mask,
sizew,
sizeh,
dtype,
scale_xy,
scale_num_inst,
use_tie_breaker,
stride,
fpn_limits=None):
"""Convert ground truth for use in loss functions.
Args:
y_true: tf.Tensor[] ground truth
[batch, box coords[0:4], classes_onehot[0:-1], best_fit_anchor_box]
mask: list of the anchor boxes choresponding to the output,
ex. [1, 2, 3] tells this layer to predict only the first 3 anchors
in the total.
size: the dimensions of this output, for regular, it progresses from
13, to 26, to 52
num_classes: `integer` for the number of classes
dtype: expected output datatype
scale_xy: A `float` to represent the amount the boxes are scaled in the
loss function.
scale_num_inst: A `float` to represent the scale at which to multiply the
number of predicted boxes by to get the number of instances to write
to the grid.
Return:
tf.Tensor[] of shape [batch, size, size, #of_anchors, 4, 1, num_classes]
"""
# unpack required components from the input ground truth
boxes = tf.cast(y_true['bbox'], dtype)
classes = tf.expand_dims(tf.cast(y_true['classes'], dtype=dtype), axis=-1)
anchors = tf.cast(y_true['best_anchors'], dtype)
ious = tf.cast(y_true['best_iou_match'], dtype)
width = tf.cast(sizew, boxes.dtype)
height = tf.cast(sizeh, boxes.dtype)
# get the number of anchor boxes used for this anchor scale
len_masks = len(mask)
# number of anchors
num_instances = tf.shape(boxes)[-2] * scale_num_inst
# rescale the x and y centers to the size of the grid [size, size]
pull_in = tf.cast(0.5 * (scale_xy - 1), boxes.dtype)
mask = tf.cast(mask, dtype=dtype)
num_reps, viable_primary, viable_alternate, viable = _get_box_locations(
anchors, mask, boxes)
# tensor arrays for tracking samples
num_written = 0
if fpn_limits is not None:
(indexes, samples,
num_written) = _write_anchor_free_grid(boxes, classes, height, width,
num_written, stride, fpn_limits)
else:
ind_val = tf.TensorArray(
tf.int32, size=0, dynamic_size=True, element_shape=[
3,
])
ind_sample = tf.TensorArray(
dtype, size=0, dynamic_size=True, element_shape=[
8,
])
if pull_in > 0.0:
(ind_val, ind_sample,
num_written) = _write_grid(viable, num_reps, boxes, classes, ious,
ind_val, ind_sample, height, width,
num_written, num_instances, pull_in)
else:
(ind_val, ind_sample,
num_written) = _write_grid(viable_primary, num_reps, boxes, classes,
ious, ind_val, ind_sample, height, width,
num_written, num_instances, 0.0)
if use_tie_breaker:
(ind_val, ind_sample,
num_written) = _write_grid(viable_alternate, num_reps, boxes, classes,
ious, ind_val, ind_sample, height, width,
num_written, num_instances, 0.0)
indexes = ind_val.stack()
samples = ind_sample.stack()
(_, ind_mask, _, _, num_reps) = tf.split(samples, [4, 1, 1, 1, 1], axis=-1)
full = tf.zeros([sizeh, sizew, len_masks, 1], dtype=dtype)
full = tf.tensor_scatter_nd_add(full, indexes, ind_mask)
if num_written >= num_instances:
tf.print("clipped")
indexs = pad_max_instances(indexes, num_instances, pad_value=0, pad_axis=0)
samples = pad_max_instances(samples, num_instances, pad_value=0, pad_axis=0)
return indexs, samples, full
def get_best_anchor(y_true,
anchors,
width=1,
height=1,
iou_thresh=0.25,
best_match_only=False):
"""
get the correct anchor that is assoiciated with each box using IOU
Args:
y_true: tf.Tensor[] for the list of bounding boxes in the yolo format
anchors: list or tensor for the anchor boxes to be used in prediction
found via Kmeans
width: int for the image width
height: int for the image height
Return:
tf.Tensor: y_true with the anchor associated with each ground truth
box known
"""
with tf.name_scope('get_best_anchor'):
is_batch = True
ytrue_shape = y_true.get_shape()
if ytrue_shape.ndims == 2:
is_batch = False
y_true = tf.expand_dims(y_true, 0)
elif ytrue_shape.ndims is None:
is_batch = False
y_true = tf.expand_dims(y_true, 0)
y_true.set_shape([None] * 3)
elif ytrue_shape.ndims != 3:
raise ValueError('\'box\' (shape %s) must have either 3 or 4 dimensions.')
width = tf.cast(width, dtype=tf.float32)
height = tf.cast(height, dtype=tf.float32)
scaler = tf.convert_to_tensor([width, height])
true_wh = tf.cast(y_true[..., 2:4], dtype=tf.float32) * scaler
anchors = tf.cast(anchors, dtype=tf.float32)
k = tf.shape(anchors)[0]
anchors = tf.expand_dims(
tf.concat([tf.zeros_like(anchors), anchors], axis=-1), axis=0)
truth_comp = tf.concat([tf.zeros_like(true_wh), true_wh], axis=-1)
if iou_thresh >= 1.0:
anchors = tf.expand_dims(anchors, axis=-2)
truth_comp = tf.expand_dims(truth_comp, axis=-3)
aspect = truth_comp[..., 2:4] / anchors[..., 2:4]
aspect = tf.where(tf.math.is_nan(aspect), tf.zeros_like(aspect), aspect)
aspect = tf.maximum(aspect, 1 / aspect)
aspect = tf.where(tf.math.is_nan(aspect), tf.zeros_like(aspect), aspect)
aspect = tf.reduce_max(aspect, axis=-1)
values, indexes = tf.math.top_k(
tf.transpose(-aspect, perm=[0, 2, 1]),
k=tf.cast(k, dtype=tf.int32),
sorted=True)
values = -values
ind_mask = tf.cast(values < iou_thresh, dtype=indexes.dtype)
else:
# iou_raw = box_ops.compute_iou(truth_comp, anchors)
truth_comp = box_ops.xcycwh_to_yxyx(truth_comp)
anchors = box_ops.xcycwh_to_yxyx(anchors)
iou_raw = box_ops.aggregated_comparitive_iou(
truth_comp,
anchors,
iou_type=3,
)
values, indexes = tf.math.top_k(
iou_raw, #tf.transpose(iou_raw, perm=[0, 2, 1]),
k=tf.cast(k, dtype=tf.int32),
sorted=True)
ind_mask = tf.cast(values >= iou_thresh, dtype=indexes.dtype)
# pad the indexs such that all values less than the thresh are -1
# add one, multiply the mask to zeros all the bad locations
# subtract 1 makeing all the bad locations 0.
if best_match_only:
iou_index = ((indexes[..., 0:] + 1) * ind_mask[..., 0:]) - 1
else:
iou_index = tf.concat([
tf.expand_dims(indexes[..., 0], axis=-1),
((indexes[..., 1:] + 1) * ind_mask[..., 1:]) - 1
],
axis=-1)
true_prod = tf.reduce_prod(true_wh, axis=-1, keepdims=True)
iou_index = tf.where(true_prod > 0, iou_index, tf.zeros_like(iou_index) - 1)
if not is_batch:
iou_index = tf.squeeze(iou_index, axis=0)
values = tf.squeeze(values, axis=0)
return tf.cast(iou_index, dtype=tf.float32), tf.cast(values, dtype=tf.float32)
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment