feat: expose logo recognition API proto for GA (#402)

packages/google-cloud-videointelligence/protos/google/cloud/videointelligence/v1/video_intelligence.proto

@@ -1,4 +1,4 @@
-// Copyright 2019 Google LLC.
+// Copyright 2020 Google LLC
 //
 // Licensed under the Apache License, Version 2.0 (the "License");
 // you may not use this file except in compliance with the License.
@@ -11,7 +11,6 @@
 // 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";
 
@@ -148,6 +147,9 @@ enum Feature {
 
   // Object detection and tracking.
   OBJECT_TRACKING = 9;
+
+  // Logo detection, tracking, and recognition.
+  LOGO_RECOGNITION = 12;
 }
 
 // Label detection mode.
@@ -397,6 +399,67 @@ message FaceAnnotation {
   repeated FaceFrame frames = 3;
 }
 
+// For tracking related features.
+// An object at time_offset with attributes, and located with
+// normalized_bounding_box.
+message TimestampedObject {
+  // Normalized Bounding box in a frame, where the object is located.
+  NormalizedBoundingBox normalized_bounding_box = 1;
+
+  // Time-offset, relative to the beginning of the video,
+  // corresponding to the video frame for this object.
+  google.protobuf.Duration time_offset = 2;
+
+  // Optional. The attributes of the object in the bounding box.
+  repeated DetectedAttribute attributes = 3 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. The detected landmarks.
+  repeated DetectedLandmark landmarks = 4 [(google.api.field_behavior) = OPTIONAL];
+}
+
+// A track of an object instance.
+message Track {
+  // Video segment of a track.
+  VideoSegment segment = 1;
+
+  // The object with timestamp and attributes per frame in the track.
+  repeated TimestampedObject timestamped_objects = 2;
+
+  // Optional. Attributes in the track level.
+  repeated DetectedAttribute attributes = 3 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. The confidence score of the tracked object.
+  float confidence = 4 [(google.api.field_behavior) = OPTIONAL];
+}
+
+// A generic detected attribute represented by name in string format.
+message DetectedAttribute {
+  // The name of the attribute, i.e. glasses, dark_glasses, mouth_open etc.
+  // A full list of supported type names will be provided in the document.
+  string name = 1;
+
+  // Detected attribute confidence. Range [0, 1].
+  float confidence = 2;
+
+  // Text value of the detection result. For example, the value for "HairColor"
+  // can be "black", "blonde", etc.
+  string value = 3;
+}
+
+// A generic detected landmark represented by name in string format and a 2D
+// location.
+message DetectedLandmark {
+  // The name of this landmark, i.e. left_hand, right_shoulder.
+  string name = 1;
+
+  // The 2D point of the detected landmark using the normalized image
+  // coordindate system. The normalized coordinates have the range from 0 to 1.
+  NormalizedVertex point = 2;
+
+  // The confidence score of the detected landmark. Range [0, 1].
+  float confidence = 3;
+}
+
 // Annotation results for a single video.
 message VideoAnnotationResults {
   // Video file location in
@@ -453,6 +516,9 @@ message VideoAnnotationResults {
   // Annotations for list of objects detected and tracked in video.
   repeated ObjectTrackingAnnotation object_annotations = 14;
 
+  // Annotations for list of logos detected, tracked and recognized in video.
+  repeated LogoRecognitionAnnotation logo_recognition_annotations = 19;
+
   // If set, indicates an error. Note that for a single `AnnotateVideoRequest`
   // some videos may succeed and some may fail.
   google.rpc.Status error = 9;
@@ -743,3 +809,18 @@ message ObjectTrackingAnnotation {
   // Streaming mode: it can only be one ObjectTrackingFrame message in frames.
   repeated ObjectTrackingFrame frames = 2;
 }
+
+// Annotation corresponding to one detected, tracked and recognized logo class.
+message LogoRecognitionAnnotation {
+  // Entity category information to specify the logo class that all the logo
+  // tracks within this LogoRecognitionAnnotation are recognized as.
+  Entity entity = 1;
+
+  // All logo tracks where the recognized logo appears. Each track corresponds
+  // to one logo instance appearing in consecutive frames.
+  repeated Track tracks = 2;
+
+  // All video segments where the recognized logo appears. There might be
+  // multiple instances of the same logo class appearing in one VideoSegment.
+  repeated VideoSegment segments = 3;
+}