diff --git a/google/cloud/vision/v1/geometry.proto b/google/cloud/vision/v1/geometry.proto
new file mode 100644
index 00000000..82c02092
--- /dev/null
+++ b/google/cloud/vision/v1/geometry.proto
@@ -0,0 +1,53 @@
+// Copyright 2016 Google Inc.
+//
+// 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.
+
+syntax = "proto3";
+
+package google.cloud.vision.v1;
+
+option cc_enable_arenas = true;
+option java_multiple_files = true;
+option java_outer_classname = "GeometryProto";
+option java_package = "com.google.cloud.vision.v1";
+
+
+// A vertex represents a 2D point in the image.
+// NOTE: the vertex coordinates are in the same scale as the original image.
+message Vertex {
+  // X coordinate.
+  int32 x = 1;
+
+  // Y coordinate.
+  int32 y = 2;
+}
+
+// A bounding polygon for the detected image annotation.
+message BoundingPoly {
+  // The bounding polygon vertices.
+  repeated Vertex vertices = 1;
+}
+
+// A 3D position in the image, used primarily for Face detection landmarks.
+// A valid Position must have both x and y coordinates.
+// The position coordinates are in the same scale as the original image.
+message Position {
+  // X coordinate.
+  float x = 1;
+
+  // Y coordinate.
+  float y = 2;
+
+  // Z coordinate (or depth).
+  float z = 3;
+}
diff --git a/google/cloud/vision/v1/image_annotator.proto b/google/cloud/vision/v1/image_annotator.proto
index ba825fcd..a25ac57b 100644
--- a/google/cloud/vision/v1/image_annotator.proto
+++ b/google/cloud/vision/v1/image_annotator.proto
@@ -18,8 +18,6 @@ package google.cloud.vision.v1;
 
 import "google/api/annotations.proto";
 import "google/cloud/vision/v1/geometry.proto";
-import "google/cloud/vision/v1/image_context_search_extension.proto";
-import "google/cloud/vision/v1/query_annotation.proto";
 import "google/rpc/status.proto";
 import "google/type/color.proto";
 import "google/type/latlng.proto";
@@ -223,7 +221,7 @@ message FaceAnnotation {
     Type type = 3;
 
     // Face landmark position.
-    google.cloud.vision.v1.Position position = 4;
+    Position position = 4;
   }
 
   // The bounding polygon around the face. The coordinates of the bounding box
@@ -233,7 +231,7 @@ message FaceAnnotation {
   // Note that one or more x and/or y coordinates may not be generated in the
   // BoundingPoly (the polygon will be unbounded) if only a partial face appears in
   // the image to be annotated.
-  google.cloud.vision.v1.BoundingPoly bounding_poly = 1;
+  BoundingPoly bounding_poly = 1;
 
   // This bounding polygon is tighter than the previous
   // <code>boundingPoly</code>, and
@@ -242,7 +240,7 @@ message FaceAnnotation {
   // "amount of skin" visible in an image. It is not based on the
   // landmarker results, only on the initial face detection, hence
   // the <code>fd</code> (face detection) prefix.
-  google.cloud.vision.v1.BoundingPoly fd_bounding_poly = 2;
+  BoundingPoly fd_bounding_poly = 2;
 
   // Detected face landmarks.
   repeated Landmark landmarks = 3;
@@ -339,7 +337,7 @@ message EntityAnnotation {
   // for `LABEL_DETECTION` features. For `TEXT_DETECTION` (OCR), `boundingPoly`s
   // are produced for the entire text detected in an image region, followed by
   // `boundingPoly`s for each word within the detected text.
-  google.cloud.vision.v1.BoundingPoly bounding_poly = 7;
+  BoundingPoly bounding_poly = 7;
 
   // The location information for the detected entity. Multiple
   // <code>LocationInfo</code> elements can be present since one location may