File size: 10,693 Bytes
5672777
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
# Copyright 2023 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.

"""Detection input and model functions for serving/inference."""

import math
from typing import Mapping, Tuple

from absl import logging
import tensorflow as tf, tf_keras

from official.vision import configs
from official.vision.modeling import factory
from official.vision.ops import anchor
from official.vision.ops import box_ops
from official.vision.ops import preprocess_ops
from official.vision.serving import export_base


class DetectionModule(export_base.ExportModule):
  """Detection Module."""

  @property
  def _padded_size(self):
    if self.params.task.train_data.parser.pad:
      return preprocess_ops.compute_padded_size(
          self._input_image_size, 2**self.params.task.model.max_level
      )
    else:
      return self._input_image_size

  def _build_model(self):

    nms_versions_supporting_dynamic_batch_size = {'batched', 'v2', 'v3'}
    nms_version = self.params.task.model.detection_generator.nms_version
    if (self._batch_size is None and
        nms_version not in nms_versions_supporting_dynamic_batch_size):
      logging.info('nms_version is set to `batched` because `%s` '
                   'does not support with dynamic batch size.', nms_version)
      self.params.task.model.detection_generator.nms_version = 'batched'

    input_specs = tf_keras.layers.InputSpec(shape=[
        self._batch_size, *self._padded_size, 3])

    if isinstance(self.params.task.model, configs.maskrcnn.MaskRCNN):
      model = factory.build_maskrcnn(
          input_specs=input_specs, model_config=self.params.task.model)
    elif isinstance(self.params.task.model, configs.retinanet.RetinaNet):
      model = factory.build_retinanet(
          input_specs=input_specs, model_config=self.params.task.model)
    else:
      raise ValueError('Detection module not implemented for {} model.'.format(
          type(self.params.task.model)))

    return model

  def _build_anchor_boxes(self):
    """Builds and returns anchor boxes."""
    model_params = self.params.task.model
    input_anchor = anchor.build_anchor_generator(
        min_level=model_params.min_level,
        max_level=model_params.max_level,
        num_scales=model_params.anchor.num_scales,
        aspect_ratios=model_params.anchor.aspect_ratios,
        anchor_size=model_params.anchor.anchor_size)
    return input_anchor(image_size=self._padded_size)

  def _build_inputs(self, image):
    """Builds detection model inputs for serving."""
    # Normalizes image with mean and std pixel values.
    image = preprocess_ops.normalize_image(
        image, offset=preprocess_ops.MEAN_RGB, scale=preprocess_ops.STDDEV_RGB)

    image, image_info = preprocess_ops.resize_and_crop_image(
        image,
        self._input_image_size,
        padded_size=self._padded_size,
        aug_scale_min=1.0,
        aug_scale_max=1.0,
        keep_aspect_ratio=self.params.task.train_data.parser.keep_aspect_ratio,
    )
    anchor_boxes = self._build_anchor_boxes()

    return image, anchor_boxes, image_info

  def _normalize_coordinates(self, detections_dict, dict_keys, image_info):
    """Normalizes detection coordinates between 0 and 1.

    Args:
      detections_dict: Dictionary containing the output of the model prediction.
      dict_keys: Key names corresponding to the tensors of the output dictionary
        that we want to update.
      image_info: Tensor containing the details of the image resizing.

    Returns:
      detections_dict: Updated detection dictionary.
    """
    for key in dict_keys:
      if key not in detections_dict:
        continue
      detection_boxes = detections_dict[key] / tf.tile(
          image_info[:, 2:3, :], [1, 1, 2]
      )
      detections_dict[key] = box_ops.normalize_boxes(
          detection_boxes, image_info[:, 0:1, :]
      )
      detections_dict[key] = tf.clip_by_value(detections_dict[key], 0.0, 1.0)

    return detections_dict

  def preprocess(
      self, images: tf.Tensor
  ) -> Tuple[tf.Tensor, Mapping[str, tf.Tensor], tf.Tensor]:
    """Preprocesses inputs to be suitable for the model.

    Args:
      images: The images tensor.
    Returns:
      images: The images tensor cast to float.
      anchor_boxes: Dict mapping anchor levels to anchor boxes.
      image_info: Tensor containing the details of the image resizing.

    """
    model_params = self.params.task.model
    with tf.device('cpu:0'):
      images = tf.cast(images, dtype=tf.float32)

      # Tensor Specs for map_fn outputs (images, anchor_boxes, and image_info).
      images_spec = tf.TensorSpec(shape=self._padded_size + [3],
                                  dtype=tf.float32)

      num_anchors = model_params.anchor.num_scales * len(
          model_params.anchor.aspect_ratios) * 4
      anchor_shapes = []
      for level in range(model_params.min_level, model_params.max_level + 1):
        anchor_level_spec = tf.TensorSpec(
            shape=[
                math.ceil(self._padded_size[0] / 2**level),
                math.ceil(self._padded_size[1] / 2**level),
                num_anchors,
            ],
            dtype=tf.float32)
        anchor_shapes.append((str(level), anchor_level_spec))

      image_info_spec = tf.TensorSpec(shape=[4, 2], dtype=tf.float32)

      images, anchor_boxes, image_info = tf.nest.map_structure(
          tf.identity,
          tf.map_fn(
              self._build_inputs,
              elems=images,
              fn_output_signature=(images_spec, dict(anchor_shapes),
                                   image_info_spec),
              parallel_iterations=32))

      return images, anchor_boxes, image_info

  def serve(self, images: tf.Tensor):
    """Casts image to float and runs inference.

    Args:
      images: uint8 Tensor of shape [batch_size, None, None, 3]
    Returns:
      Tensor holding detection output logits.
    """

    # Skip image preprocessing when input_type is tflite so it is compatible
    # with TFLite quantization.
    if self._input_type != 'tflite':
      images, anchor_boxes, image_info = self.preprocess(images)
    else:
      with tf.device('cpu:0'):
        anchor_boxes = self._build_anchor_boxes()
        # image_info is a 3D tensor of shape [batch_size, 4, 2]. It is in the
        # format of [[original_height, original_width],
        # [desired_height, desired_width], [y_scale, x_scale],
        # [y_offset, x_offset]]. When input_type is tflite, input image is
        # supposed to be preprocessed already.
        image_info = tf.convert_to_tensor([[
            self._input_image_size, self._input_image_size, [1.0, 1.0], [0, 0]
        ]],
                                          dtype=tf.float32)
    input_image_shape = image_info[:, 1, :]

    # To overcome keras.Model extra limitation to save a model with layers that
    # have multiple inputs, we use `model.call` here to trigger the forward
    # path. Note that, this disables some keras magics happens in `__call__`.
    model_call_kwargs = {
        'images': images,
        'image_shape': input_image_shape,
        'anchor_boxes': anchor_boxes,
        'training': False,
    }
    if isinstance(self.params.task.model, configs.retinanet.RetinaNet):
      model_call_kwargs['output_intermediate_features'] = (
          self.params.task.export_config.output_intermediate_features
      )
    detections = self.model.call(**model_call_kwargs)

    if self.params.task.model.detection_generator.apply_nms:
      # For RetinaNet model, apply export_config.
      # TODO(huizhongc): Add export_config to fasterrcnn and maskrcnn as needed.
      if isinstance(self.params.task.model, configs.retinanet.RetinaNet):
        export_config = self.params.task.export_config
        # Normalize detection box coordinates to [0, 1].
        if export_config.output_normalized_coordinates:
          keys = ['detection_boxes', 'detection_outer_boxes']
          detections = self._normalize_coordinates(detections, keys, image_info)

        # Cast num_detections and detection_classes to float. This allows the
        # model inference to work on chain (go/chain) as chain requires floating
        # point outputs.
        if export_config.cast_num_detections_to_float:
          detections['num_detections'] = tf.cast(
              detections['num_detections'], dtype=tf.float32)
        if export_config.cast_detection_classes_to_float:
          detections['detection_classes'] = tf.cast(
              detections['detection_classes'], dtype=tf.float32)

      final_outputs = {
          'detection_boxes': detections['detection_boxes'],
          'detection_scores': detections['detection_scores'],
          'detection_classes': detections['detection_classes'],
          'num_detections': detections['num_detections']
      }
      if 'detection_outer_boxes' in detections:
        final_outputs['detection_outer_boxes'] = (
            detections['detection_outer_boxes'])
    else:
      # For RetinaNet model, apply export_config.
      if isinstance(self.params.task.model, configs.retinanet.RetinaNet):
        export_config = self.params.task.export_config
        # Normalize detection box coordinates to [0, 1].
        if export_config.output_normalized_coordinates:
          keys = ['decoded_boxes']
          detections = self._normalize_coordinates(detections, keys, image_info)
      final_outputs = {
          'decoded_boxes': detections['decoded_boxes'],
          'decoded_box_scores': detections['decoded_box_scores']
      }

    if 'detection_masks' in detections.keys():
      final_outputs['detection_masks'] = detections['detection_masks']
    if (
        isinstance(self.params.task.model, configs.retinanet.RetinaNet)
        and self.params.task.export_config.output_intermediate_features
    ):
      final_outputs.update(
          {
              k: v
              for k, v in detections.items()
              if k.startswith('backbone_') or k.startswith('decoder_')
          }
      )

    if self.params.task.model.detection_generator.nms_version != 'tflite':
      final_outputs.update({'image_info': image_info})
    return final_outputs