add option to generate panoptic masks

official/vision/beta/projects/panoptic_maskrcnn/modeling/factory.py

@@ -222,15 +222,20 @@ def build_panoptic_deeplab(
       norm_epsilon=norm_activation_config.norm_epsilon,
       kernel_regularizer=l2_regularizer)
 
+  if model_config.generate_panoptic_masks:
     post_processing_config = model_config.post_processor
     post_processor = panoptic_deeplab_merge.PostProcessor(
+        output_size=post_processing_config.output_size,
         center_score_threshold=post_processing_config.center_score_threshold,
         thing_class_ids=post_processing_config.thing_class_ids,
         label_divisor=post_processing_config.label_divisor,
         stuff_area_limit=post_processing_config.stuff_area_limit,
         ignore_label=post_processing_config.ignore_label,
         nms_kernel=post_processing_config.nms_kernel,
-      keep_k_centers=post_processing_config.keep_k_centers)
+        keep_k_centers=post_processing_config.keep_k_centers,
+        rescale_predictions=post_processing_config.rescale_predictions)
+  else:
+    post_processor = None
 
   model = panoptic_deeplab_model.PanopticDeeplabModel(
       backbone=backbone, 

official/vision/beta/projects/panoptic_maskrcnn/modeling/factory_test.py

@@ -73,9 +73,11 @@ class PanopticDeeplabBuilderTest(parameterized.TestCase, tf.test.TestCase):
           decoder_type=['aspp', 'fpn'],
           level=[2, 3, 4],
           low_level=[(4, 3), (3, 2)],
-          shared_decoder=[True, False]))
-  def test_builder(self, input_size, backbone_type, level, 
-                   low_level, decoder_type, shared_decoder):
+          shared_decoder=[True, False],
+          generate_panoptic_masks=[True, False]))
+  def test_builder(self, input_size, backbone_type,
+                   level, low_level, decoder_type,
+                   shared_decoder, generate_panoptic_masks):
     num_classes = 10
     input_specs = tf.keras.layers.InputSpec(
         shape=[None, input_size[0], input_size[1], 3])
@@ -97,7 +99,8 @@ class PanopticDeeplabBuilderTest(parameterized.TestCase, tf.test.TestCase):
             kernel_size=5,
             prediction_kernel_size=1,
             low_level=low_level),
-        shared_decoder=shared_decoder)
+        shared_decoder=shared_decoder,
+        generate_panoptic_masks=generate_panoptic_masks)
 
     l2_regularizer = tf.keras.regularizers.l2(5e-5)
     _ = factory.build_panoptic_deeplab(

official/vision/beta/projects/panoptic_maskrcnn/modeling/panoptic_deeplab_model.py

@@ -57,7 +57,10 @@ class PanopticDeeplabModel(tf.keras.Model):
     self.instance_head = instance_head
     self.post_processor = post_processor
 
-  def call(self, inputs: tf.Tensor, training: bool = None) -> tf.Tensor:
+  def call(
+      self, inputs: tf.Tensor,
+      image_info: tf.Tensor,
+      training: bool = None) -> tf.Tensor:
     if training is None:
       training = tf.keras.backend.learning_phase()
 
@@ -81,15 +84,18 @@ class PanopticDeeplabModel(tf.keras.Model):
 
     outputs = {
         'segmentation_outputs': segmentation_outputs,
-        'instance_center_prediction': 
-            instance_outputs['instance_center_prediction'],
-        'instance_center_regression': 
-            instance_outputs['instance_center_regression'],
+        'instance_centers_heatmap': 
+            instance_outputs['instance_centers_heatmap'],
+        'instance_centers_offset': 
+            instance_outputs['instance_centers_offset'],
     }
     if training:
       return outputs
 
-    outputs = self.post_processor(outputs)
+    if self.post_processor is not None:
+      panoptic_masks = self.post_processor(outputs, image_info)
+      outputs.update(panoptic_masks)
+      
     return outputs
 
   @property