Project import generated by Copybara.

GitOrigin-RevId: d91373b4d4d10abef49cab410caa6aadf0875049

mediapipe/graphs/object_detection/object_detection_desktop_lstd_tflite_graph.pbtxt

@@ -0,0 +1,176 @@
+# MediaPipe graph that performs LSTM based object detection on desktop with
+# TensorFlow Lite on CPU.
+
+# max_queue_size limits the number of packets enqueued on any input stream
+# by throttling inputs to the graph. This makes the graph only process one
+# frame per time.
+max_queue_size: 1
+
+# Decodes an input video file into images and a video header.
+node {
+  calculator: "OpenCvVideoDecoderCalculator"
+  input_side_packet: "INPUT_FILE_PATH:input_video_path"
+  output_stream: "VIDEO:input_video"
+  output_stream: "VIDEO_PRESTREAM:input_video_header"
+}
+
+# Transforms the input image on CPU to a 256x256 image. To scale the image, by
+# default it uses the STRETCH scale mode that maps the entire input image to the
+# entire transformed image. As a result, image aspect ratio may be changed and
+# objects in the image may be deformed (stretched or squeezed), but the object
+# detection model used in this graph is agnostic to that deformation.
+node: {
+  calculator: "ImageTransformationCalculator"
+  input_stream: "IMAGE:input_video"
+  output_stream: "IMAGE:transformed_input_video"
+  node_options: {
+    [type.googleapis.com/mediapipe.ImageTransformationCalculatorOptions] {
+      output_width: 256
+      output_height: 256
+    }
+  }
+}
+
+# Converts the transformed input image on CPU into an image tensor as a
+# TfLiteTensor. The zero_center option is set to true to normalize the
+# pixel values to [-1.f, 1.f] as opposed to [0.f, 1.f].
+node {
+  calculator: "TfLiteConverterCalculator"
+  input_stream: "IMAGE:transformed_input_video"
+  output_stream: "TENSORS:image_tensor"
+  node_options: {
+    [type.googleapis.com/mediapipe.TfLiteConverterCalculatorOptions] {
+      zero_center: true
+      use_quantized_tensors: true
+    }
+  }
+}
+
+# Joins the input image_tensor and previous lstm state tensors into a single
+# output vector of tensors.
+node {
+  calculator: "ConcatenateTfLiteTensorVectorCalculator"
+  input_stream: "image_tensor"
+  input_stream: "prev_lstm_tensors"
+  output_stream: "concatenated_tensors"
+}
+
+# Runs a TensorFlow Lite model on CPU that takes an image tensor and state
+# tensors, then outputs a vector of tensors representing, for instance,
+# detection boxes/keypoints and scores.
+node {
+  calculator: "TfLiteInferenceCalculator"
+  input_stream: "TENSORS:concatenated_tensors"
+  output_stream: "TENSORS:inference_tensors"
+  node_options: {
+    [type.googleapis.com/mediapipe.TfLiteInferenceCalculatorOptions] {
+      model_path: "mediapipe/models/lstdlite_object_detection.tflite"
+    }
+  }
+}
+
+# Splits up the output tensors into two streams [detection_boxes,
+# detection_classes, detection_scores, num_detections], and [lstm_c, lstm_h].
+node {
+  calculator: "SplitTfLiteTensorVectorCalculator"
+  input_stream: "inference_tensors"
+  output_stream: "detection_tensors"
+  output_stream: "lstm_tensors"
+  options {
+    [type.googleapis.com/mediapipe.SplitVectorCalculatorOptions] {
+      ranges: { begin: 0 end: 4 }
+      ranges: { begin: 4 end: 6 }
+    }
+  }
+}
+
+# Waits for the LSTM state output from the TfLite inferencing to feed back in as
+# an input tensor for the next frame.
+node {
+  calculator: "PreviousLoopbackCalculator"
+  input_stream: "MAIN:transformed_input_video"
+  input_stream: "LOOP:lstm_tensors"
+  input_stream_info: { tag_index: "LOOP" back_edge: true }
+  output_stream: "PREV_LOOP:prev_lstm_tensors"
+}
+
+# Decodes the detection tensors generated by the TensorFlow Lite model, based on
+# the specification in the options, into a vector of detections. Each detection
+# describes a detected object.
+node {
+  calculator: "TfLiteTensorsToDetectionsCalculator"
+  input_stream: "TENSORS:detection_tensors"
+  output_stream: "DETECTIONS:detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.TfLiteTensorsToDetectionsCalculatorOptions] {
+      num_classes: 7
+      num_boxes: 40
+      num_coords: 4
+    }
+  }
+}
+
+# Performs non-max suppression to remove excessive detections.
+node {
+  calculator: "NonMaxSuppressionCalculator"
+  input_stream: "detections"
+  output_stream: "filtered_detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.NonMaxSuppressionCalculatorOptions] {
+      min_suppression_threshold: 0.4
+      min_score_threshold: 0.2
+      max_num_detections: 5
+      overlap_type: INTERSECTION_OVER_UNION
+    }
+  }
+}
+
+# Maps detection label IDs to the corresponding label text. The label map is
+# provided in the label_map_path option.
+node {
+  calculator: "DetectionLabelIdToTextCalculator"
+  input_stream: "filtered_detections"
+  output_stream: "output_detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.DetectionLabelIdToTextCalculatorOptions] {
+      label_map_path: "mediapipe/models/lstdlite_object_detection_labelmap.txt"
+    }
+  }
+}
+
+# Converts the detections to drawing primitives for annotation overlay.
+node {
+  calculator: "DetectionsToRenderDataCalculator"
+  input_stream: "DETECTIONS:output_detections"
+  output_stream: "RENDER_DATA:render_data"
+  node_options: {
+    [type.googleapis.com/mediapipe.DetectionsToRenderDataCalculatorOptions] {
+      thickness: 2.0
+      color { r: 255 g: 255 b: 255 }
+    }
+  }
+}
+
+# Draws annotations and overlays them on top of the original image coming into
+# the graph.
+node {
+  calculator: "AnnotationOverlayCalculator"
+  input_stream: "INPUT_FRAME:input_video"
+  input_stream: "render_data"
+  output_stream: "OUTPUT_FRAME:output_video"
+}
+
+# Encodes the annotated images into a video file, adopting properties specified
+# in the input video header, e.g., video framerate.
+node {
+  calculator: "OpenCvVideoEncoderCalculator"
+  input_stream: "VIDEO:output_video"
+  input_stream: "VIDEO_PRESTREAM:input_video_header"
+  input_side_packet: "OUTPUT_FILE_PATH:output_video_path"
+  node_options: {
+    [type.googleapis.com/mediapipe.OpenCvVideoEncoderCalculatorOptions]: {
+      codec: "avc1"
+      video_format: "mp4"
+    }
+  }
+}

mediapipe/graphs/tracking/BUILD

@@ -0,0 +1,39 @@
+# Copyright 2019 The MediaPipe Authors.
+#
+# 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.
+
+load(
+    "//mediapipe/framework/tool:mediapipe_graph.bzl",
+    "mediapipe_binary_graph",
+)
+
+licenses(["notice"])  # Apache 2.0
+
+package(default_visibility = ["//visibility:public"])
+
+cc_library(
+    name = "mobile_calculators",
+    deps = [
+        "//mediapipe/calculators/core:packet_resampler_calculator",
+        "//mediapipe/graphs/tracking/subgraphs:object_detection_gpu",
+        "//mediapipe/graphs/tracking/subgraphs:object_tracking",
+        "//mediapipe/graphs/tracking/subgraphs:renderer_gpu",
+    ],
+)
+
+mediapipe_binary_graph(
+    name = "mobile_gpu_binary_graph",
+    graph = "object_detection_tracking_mobile_gpu.pbtxt",
+    output_name = "mobile_gpu.binarypb",
+    deps = [":mobile_calculators"],
+)

mediapipe/graphs/tracking/object_detection_tracking_mobile_gpu.pbtxt

@@ -0,0 +1,46 @@
+# MediaPipe graph that performs object detection and tracking with TensorFlow
+# Lite on GPU.
+# Used in the examples in
+# mediapipie/examples/android/src/java/com/mediapipe/apps/objecttrackinggpu
+
+# Images on GPU coming into and out of the graph.
+input_stream: "input_video"
+output_stream: "output_video"
+
+# Resamples the images by specific frame rate. This calculator is used to
+# control the frequecy of subsequent calculators/subgraphs, e.g. less power
+# consumption for expensive process.
+node {
+  calculator: "PacketResamplerCalculator"
+  input_stream: "DATA:input_video"
+  output_stream: "DATA:throttled_input_video"
+  node_options: {
+    [type.googleapis.com/mediapipe.PacketResamplerCalculatorOptions] {
+      frame_rate: 0.5
+    }
+  }
+}
+
+# Subgraph that detections objects (see object_detection_gpu.pbtxt).
+node {
+  calculator: "ObjectDetectionSubgraph"
+  input_stream: "IMAGE:throttled_input_video"
+  output_stream: "DETECTIONS:output_detections"
+}
+
+# Subgraph that tracks objects (see object_tracking.pbtxt).
+node {
+  calculator: "ObjectTrackingSubgraph"
+  input_stream: "VIDEO:input_video"
+  input_stream: "DETECTIONS:output_detections"
+  output_stream: "DETECTIONS:tracked_detections"
+}
+
+# Subgraph that renders annotations and overlays them on top of the input
+# images (see renderer_gpu.pbtxt).
+node {
+  calculator: "RendererSubgraph"
+  input_stream: "IMAGE:input_video"
+  input_stream: "DETECTIONS:tracked_detections"
+  output_stream: "IMAGE:output_video"
+}

mediapipe/graphs/tracking/subgraphs/BUILD

@@ -0,0 +1,76 @@
+# Copyright 2019 The MediaPipe Authors.
+#
+# 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.
+
+load(
+    "//mediapipe/framework/tool:mediapipe_graph.bzl",
+    "mediapipe_simple_subgraph",
+)
+
+licenses(["notice"])  # Apache 2.0
+
+package(default_visibility = ["//visibility:public"])
+
+mediapipe_simple_subgraph(
+    name = "box_tracking",
+    graph = "box_tracking.pbtxt",
+    register_as = "BoxTrackingSubgraph",
+    deps = [
+        "//mediapipe/calculators/image:image_transformation_calculator",
+        "//mediapipe/calculators/video:box_tracker_calculator",
+        "//mediapipe/calculators/video:flow_packager_calculator",
+        "//mediapipe/calculators/video:motion_analysis_calculator",
+        "//mediapipe/framework/stream_handler:immediate_input_stream_handler",
+        "//mediapipe/framework/stream_handler:sync_set_input_stream_handler",
+        "//mediapipe/gpu:gpu_buffer_to_image_frame_calculator",
+    ],
+)
+
+mediapipe_simple_subgraph(
+    name = "object_tracking",
+    graph = "object_tracking.pbtxt",
+    register_as = "ObjectTrackingSubgraph",
+    deps = [
+        "//mediapipe/calculators/util:detection_unique_id_calculator",
+        "//mediapipe/calculators/util:detections_to_timed_box_list_calculator",
+        "//mediapipe/calculators/video:tracked_detection_manager_calculator",
+        "//mediapipe/framework/stream_handler:sync_set_input_stream_handler",
+        "//mediapipe/graphs/tracking/subgraphs:box_tracking",
+    ],
+)
+
+mediapipe_simple_subgraph(
+    name = "object_detection_gpu",
+    graph = "object_detection_gpu.pbtxt",
+    register_as = "ObjectDetectionSubgraph",
+    deps = [
+        "//mediapipe/calculators/image:image_transformation_calculator",
+        "//mediapipe/calculators/tflite:ssd_anchors_calculator",
+        "//mediapipe/calculators/tflite:tflite_converter_calculator",
+        "//mediapipe/calculators/tflite:tflite_inference_calculator",
+        "//mediapipe/calculators/tflite:tflite_tensors_to_detections_calculator",
+        "//mediapipe/calculators/util:detection_label_id_to_text_calculator",
+        "//mediapipe/calculators/util:non_max_suppression_calculator",
+    ],
+)
+
+mediapipe_simple_subgraph(
+    name = "renderer_gpu",
+    graph = "renderer_gpu.pbtxt",
+    register_as = "RendererSubgraph",
+    deps = [
+        "//mediapipe/calculators/util:annotation_overlay_calculator",
+        "//mediapipe/calculators/util:detections_to_render_data_calculator",
+        "//mediapipe/calculators/util:rect_to_render_data_calculator",
+    ],
+)

mediapipe/graphs/tracking/subgraphs/box_tracking.pbtxt

@@ -0,0 +1,126 @@
+# MediaPipe box tracking subgraph.
+
+type: "BoxTrackingSubgraph"
+
+input_stream: "VIDEO:input_video"
+input_stream: "BOXES:start_pos"
+input_stream: "CANCEL_ID:cancel_object_id"
+output_stream: "BOXES:boxes"
+
+node: {
+  calculator: "ImageTransformationCalculator"
+  input_stream: "IMAGE_GPU:input_video"
+  output_stream: "IMAGE_GPU:downscaled_input_video"
+  node_options: {
+    [type.googleapis.com/mediapipe.ImageTransformationCalculatorOptions] {
+      output_width: 240
+      output_height: 320
+    }
+  }
+}
+
+# Converts GPU buffer to ImageFrame for processing tracking.
+node: {
+  calculator: "GpuBufferToImageFrameCalculator"
+  input_stream: "downscaled_input_video"
+  output_stream: "downscaled_input_video_cpu"
+}
+
+# Performs motion analysis on an incoming video stream.
+node: {
+  calculator: "MotionAnalysisCalculator"
+  input_stream: "VIDEO:downscaled_input_video_cpu"
+  output_stream: "CAMERA:camera_motion"
+  output_stream: "FLOW:region_flow"
+
+  node_options: {
+    [type.googleapis.com/mediapipe.MotionAnalysisCalculatorOptions]: {
+      analysis_options {
+        analysis_policy: ANALYSIS_POLICY_CAMERA_MOBILE
+        flow_options {
+          fast_estimation_min_block_size: 100
+          top_inlier_sets: 1
+          frac_inlier_error_threshold: 3e-3
+          downsample_mode: DOWNSAMPLE_TO_INPUT_SIZE
+          verification_distance: 5.0
+          verify_long_feature_acceleration: true
+          verify_long_feature_trigger_ratio: 0.1
+          tracking_options {
+            max_features: 500
+            adaptive_extraction_levels: 2
+            min_eig_val_settings {
+              adaptive_lowest_quality_level: 2e-4
+            }
+            klt_tracker_implementation: KLT_OPENCV
+          }
+        }
+      }
+    }
+  }
+}
+
+# Reads optical flow fields defined in
+# mediapipe/framework/formats/motion/optical_flow_field.h,
+# returns a VideoFrame with 2 channels (v_x and v_y), each channel is quantized
+# to 0-255.
+node: {
+  calculator: "FlowPackagerCalculator"
+  input_stream: "FLOW:region_flow"
+  input_stream: "CAMERA:camera_motion"
+  output_stream: "TRACKING:tracking_data"
+
+  node_options: {
+    [type.googleapis.com/mediapipe.FlowPackagerCalculatorOptions]: {
+      flow_packager_options: {
+        binary_tracking_data_support: false
+      }
+    }
+  }
+}
+
+# Tracks box positions over time.
+node: {
+  calculator: "BoxTrackerCalculator"
+  input_stream: "TRACKING:tracking_data"
+  input_stream: "TRACK_TIME:input_video"
+  input_stream: "START_POS:start_pos"
+  input_stream: "CANCEL_OBJECT_ID:cancel_object_id"
+  input_stream_info: {
+    tag_index: "CANCEL_OBJECT_ID"
+    back_edge: true
+  }
+  output_stream: "BOXES:boxes"
+
+  input_stream_handler {
+    input_stream_handler: "SyncSetInputStreamHandler"
+    options {
+      [mediapipe.SyncSetInputStreamHandlerOptions.ext] {
+        sync_set {
+          tag_index: "TRACKING"
+          tag_index: "TRACK_TIME"
+        }
+        sync_set {
+          tag_index: "START_POS"
+        }
+        sync_set {
+          tag_index: "CANCEL_OBJECT_ID"
+        }
+      }
+    }
+  }
+
+  node_options: {
+    [type.googleapis.com/mediapipe.BoxTrackerCalculatorOptions]: {
+      tracker_options: {
+        track_step_options {
+          track_object_and_camera: true
+          tracking_degrees: TRACKING_DEGREE_OBJECT_SCALE
+          inlier_spring_force: 0.0
+          static_motion_temporal_ratio: 3e-2
+        }
+      }
+      visualize_tracking_data: false
+      streaming_track_data_cache_size: 100
+    }
+  }
+}

mediapipe/graphs/tracking/subgraphs/object_detection_gpu.pbtxt

@@ -0,0 +1,128 @@
+# MediaPipe object detection subgraph.
+
+type: "ObjectDetectionSubgraph"
+
+input_stream: "IMAGE:input_video"
+output_stream: "DETECTIONS:output_detections"
+
+# Transforms the input image on GPU to a 320x320 image. To scale the image, by
+# default it uses the STRETCH scale mode that maps the entire input image to the
+# entire transformed image. As a result, image aspect ratio may be changed and
+# objects in the image may be deformed (stretched or squeezed), but the object
+# detection model used in this graph is agnostic to that deformation.
+node: {
+  calculator: "ImageTransformationCalculator"
+  input_stream: "IMAGE_GPU:input_video"
+  output_stream: "IMAGE_GPU:transformed_input_video"
+  node_options: {
+    [type.googleapis.com/mediapipe.ImageTransformationCalculatorOptions] {
+      output_width: 320
+      output_height: 320
+    }
+  }
+}
+
+# Converts the transformed input image on GPU into an image tensor stored as a
+# TfLiteTensor.
+node {
+  calculator: "TfLiteConverterCalculator"
+  input_stream: "IMAGE_GPU:transformed_input_video"
+  output_stream: "TENSORS_GPU:image_tensor"
+}
+
+# Runs a TensorFlow Lite model on GPU that takes an image tensor and outputs a
+# vector of tensors representing, for instance, detection boxes/keypoints and
+# scores.
+node {
+  calculator: "TfLiteInferenceCalculator"
+  input_stream: "TENSORS_GPU:image_tensor"
+  output_stream: "TENSORS_GPU:detection_tensors"
+  node_options: {
+    [type.googleapis.com/mediapipe.TfLiteInferenceCalculatorOptions] {
+      model_path: "mediapipe/models/ssdlite_object_detection.tflite"
+    }
+  }
+}
+
+# Generates a single side packet containing a vector of SSD anchors based on
+# the specification in the options.
+node {
+  calculator: "SsdAnchorsCalculator"
+  output_side_packet: "anchors"
+  node_options: {
+    [type.googleapis.com/mediapipe.SsdAnchorsCalculatorOptions] {
+      num_layers: 6
+      min_scale: 0.2
+      max_scale: 0.95
+      input_size_height: 320
+      input_size_width: 320
+      anchor_offset_x: 0.5
+      anchor_offset_y: 0.5
+      strides: 16
+      strides: 32
+      strides: 64
+      strides: 128
+      strides: 256
+      strides: 512
+      aspect_ratios: 1.0
+      aspect_ratios: 2.0
+      aspect_ratios: 0.5
+      aspect_ratios: 3.0
+      aspect_ratios: 0.3333
+      reduce_boxes_in_lowest_layer: true
+    }
+  }
+}
+
+# Decodes the detection tensors generated by the TensorFlow Lite model, based on
+# the SSD anchors and the specification in the options, into a vector of
+# detections. Each detection describes a detected object.
+node {
+  calculator: "TfLiteTensorsToDetectionsCalculator"
+  input_stream: "TENSORS_GPU:detection_tensors"
+  input_side_packet: "ANCHORS:anchors"
+  output_stream: "DETECTIONS:detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.TfLiteTensorsToDetectionsCalculatorOptions] {
+      num_classes: 91
+      num_boxes: 2034
+      num_coords: 4
+      ignore_classes: 0
+      sigmoid_score: true
+      apply_exponential_on_box_size: true
+      x_scale: 10.0
+      y_scale: 10.0
+      h_scale: 5.0
+      w_scale: 5.0
+      min_score_thresh: 0.6
+    }
+  }
+}
+
+# Performs non-max suppression to remove excessive detections.
+node {
+  calculator: "NonMaxSuppressionCalculator"
+  input_stream: "detections"
+  output_stream: "filtered_detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.NonMaxSuppressionCalculatorOptions] {
+      min_suppression_threshold: 0.4
+      max_num_detections: 3
+      overlap_type: INTERSECTION_OVER_UNION
+      return_empty_detections: true
+    }
+  }
+}
+
+# Maps detection label IDs to the corresponding label text. The label map is
+# provided in the label_map_path option.
+node {
+  calculator: "DetectionLabelIdToTextCalculator"
+  input_stream: "filtered_detections"
+  output_stream: "output_detections"
+  node_options: {
+    [type.googleapis.com/mediapipe.DetectionLabelIdToTextCalculatorOptions] {
+      label_map_path: "mediapipe/models/ssdlite_object_detection_labelmap.txt"
+    }
+  }
+}

mediapipe/graphs/tracking/subgraphs/object_tracking.pbtxt

@@ -0,0 +1,56 @@
+# MediaPipe object tracking subgraph.
+
+type: "ObjectTrackingSubgraph"
+
+input_stream: "VIDEO:input_video"
+input_stream: "DETECTIONS:new_detections"
+output_stream: "DETECTIONS:tracked_detections"
+
+# Assigns an unique id for each new detection.
+node {
+  calculator: "DetectionUniqueIdCalculator"
+  input_stream: "DETECTIONS:new_detections"
+  output_stream: "DETECTIONS:detections_with_id"
+}
+
+# Converts detections to TimedBox protos which are used as initial location
+# for tracking.
+node {
+  calculator: "DetectionsToTimedBoxListCalculator"
+  input_stream: "DETECTIONS:detections_with_id"
+  output_stream: "BOXES:start_pos"
+}
+
+# Subgraph that tracks boxes (see box_tracking.pbtxt).
+node {
+  calculator: "BoxTrackingSubgraph"
+  input_stream: "VIDEO:input_video"
+  input_stream: "BOXES:start_pos"
+  input_stream: "CANCEL_ID:cancel_object_id"
+  output_stream: "BOXES:boxes"
+}
+
+# Managers new detected objects and objects that are being tracked.
+# It associates the duplicated detections and updates the locations of
+# detections from tracking.
+node: {
+  calculator: "TrackedDetectionManagerCalculator"
+  input_stream: "DETECTIONS:detections_with_id"
+  input_stream: "TRACKING_BOXES:boxes"
+  output_stream: "DETECTIONS:tracked_detections"
+  output_stream: "CANCEL_OBJECT_ID:cancel_object_id"
+
+  input_stream_handler {
+    input_stream_handler: "SyncSetInputStreamHandler"
+    options {
+      [mediapipe.SyncSetInputStreamHandlerOptions.ext] {
+        sync_set {
+          tag_index: "TRACKING_BOXES"
+        }
+        sync_set {
+          tag_index: "DETECTIONS"
+        }
+      }
+    }
+  }
+}

mediapipe/graphs/tracking/subgraphs/renderer_gpu.pbtxt

@@ -0,0 +1,29 @@
+# MediaPipe object tracking rendering subgraph.
+
+type: "RendererSubgraph"
+
+input_stream: "IMAGE:input_image"
+input_stream: "DETECTIONS:detections"
+output_stream: "IMAGE:output_image"
+
+# Converts the detections to drawing primitives for annotation overlay.
+node {
+  calculator: "DetectionsToRenderDataCalculator"
+  input_stream: "DETECTIONS:detections"
+  output_stream: "RENDER_DATA:detections_render_data"
+  node_options: {
+    [type.googleapis.com/mediapipe.DetectionsToRenderDataCalculatorOptions] {
+      thickness: 4.0
+      color { r: 255 g: 0 b: 0 }
+      render_detection_id: true
+    }
+  }
+}
+
+# Draws annotations and overlays them on top of the input images.
+node {
+  calculator: "AnnotationOverlayCalculator"
+  input_stream: "INPUT_FRAME_GPU:input_image"
+  input_stream: "detections_render_data"
+  output_stream: "OUTPUT_FRAME_GPU:output_image"
+}