Ensure that CenterNetMetaArch can process keypoints even when no instances are present.

PiperOrigin-RevId: 399303089

research/object_detection/meta_architectures/center_net_meta_arch.py

@@ -1585,6 +1585,13 @@ def _gather_candidates_at_indices(keypoint_candidates,
 
   gathered_keypoint_candidates = tf.transpose(
       nearby_candidate_coords_transposed, [0, 2, 1, 3])
+  # The reshape operation above may result in a singleton last dimension, but
+  # downstream code requires it to always be at least 2-valued.
+  original_shape = tf.shape(gathered_keypoint_candidates)
+  new_shape = tf.concat((original_shape[:3],
+                         [tf.maximum(original_shape[3], 2)]), 0)
+  gathered_keypoint_candidates = tf.reshape(gathered_keypoint_candidates,
+                                            new_shape)
   gathered_keypoint_scores = tf.transpose(nearby_candidate_scores_transposed,
                                           [0, 2, 1])
 

research/object_detection/meta_architectures/center_net_meta_arch_tf2_test.py

@@ -1242,6 +1242,44 @@ class CenterNetMetaArchHelpersTest(test_case.TestCase, parameterized.TestCase):
     np.testing.assert_allclose(expected_refined_keypoints, refined_keypoints)
     np.testing.assert_allclose(expected_refined_scores, refined_scores)
 
+  def test_refine_keypoints_with_empty_regressed_keypoints(self):
+    regressed_keypoints_np = np.zeros((1, 0, 2, 2), dtype=np.float32)
+    keypoint_candidates_np = np.ones((1, 1, 2, 2), dtype=np.float32)
+    keypoint_scores_np = np.ones((1, 1, 2), dtype=np.float32)
+    num_keypoints_candidates_np = np.ones((1, 1), dtype=np.int32)
+    unmatched_keypoint_score = 0.1
+
+    def graph_fn():
+      regressed_keypoints = tf.constant(
+          regressed_keypoints_np, dtype=tf.float32)
+      keypoint_candidates = tf.constant(
+          keypoint_candidates_np, dtype=tf.float32)
+      keypoint_scores = tf.constant(keypoint_scores_np, dtype=tf.float32)
+      num_keypoint_candidates = tf.constant(num_keypoints_candidates_np,
+                                            dtype=tf.int32)
+      # The behavior of bboxes=None is different now. We provide the bboxes
+      # explicitly by using the regressed keypoints to create the same
+      # behavior.
+      regressed_keypoints_flattened = tf.reshape(
+          regressed_keypoints, [-1, 3, 2])
+      bboxes_flattened = keypoint_ops.keypoints_to_enclosing_bounding_boxes(
+          regressed_keypoints_flattened)
+      (refined_keypoints, refined_scores, _) = cnma.refine_keypoints(
+          regressed_keypoints,
+          keypoint_candidates,
+          keypoint_scores,
+          num_keypoint_candidates,
+          bboxes=bboxes_flattened,
+          unmatched_keypoint_score=unmatched_keypoint_score,
+          box_scale=1.2,
+          candidate_search_scale=0.3,
+          candidate_ranking_mode='min_distance')
+      return refined_keypoints, refined_scores
+
+    refined_keypoints, refined_scores = self.execute(graph_fn, [])
+    self.assertEqual(refined_keypoints.shape, (1, 0, 2, 2))
+    self.assertEqual(refined_scores.shape, (1, 0, 2))
+
   def test_refine_keypoints_without_bbox(self):
     regressed_keypoints_np = np.array(
         [