talksik/mediapipe
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Project import generated by Copybara.

Browse Source 
PiperOrigin-RevId: 264105834
This commit is contained in:
MediaPipe Team
2019-08-19 08:50:18 +00:00
committed by Camillo Lugaresi
parent 71a47bb18b
commit f5df228d9b
28 changed files with 276 additions and 257 deletions
Download Patch File Download Diff File Expand all files Collapse all files
mediapipe/docs/hair_segmentation_mobile_gpu.md
+16 -42
View File
@@ -8,20 +8,12 @@ that performs hair segmentation with TensorFlow Lite on GPU.
## Android
Please see [Hello World! in MediaPipe on Android](hello_world_android.md) for
general instructions to develop an Android application that uses MediaPipe.
[Source](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/hairsegmentationgpu)
The graph below is used in the
[Hair Segmentation GPU Android example app](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/hairsegmentationgpu).
To build the app, run:
To build and install the app:
```bash
bazel build -c opt --config=android_arm64 mediapipe/examples/android/src/java/com/google/mediapipe/apps/hairsegmentationgpu
```
To further install the app on an Android device, run:
```bash
adb install bazel-bin/mediapipe/examples/android/src/java/com/google/mediapipe/apps/hairsegmentationgpu/hairsegmentationgpu.apk
```
@@ -37,7 +29,7 @@ below and paste it into [MediaPipe Visualizer](https://viz.mediapipe.dev/).
```bash
# MediaPipe graph that performs hair segmentation with TensorFlow Lite on GPU.
# Used in the example in
# mediapipie/examples/ios/hairsegmentationgpu.
# mediapipie/examples/android/src/java/com/mediapipe/apps/hairsegmentationgpu.
# Images on GPU coming into and out of the graph.
input_stream: "input_video"
@@ -84,14 +76,11 @@ node: {
}
}
# Waits for a mask from the previous round of hair segmentation to be fed back
# as an input, and caches it. Upon the arrival of an input image, it checks if
# there is a mask cached, and sends out the mask with the timestamp replaced by
# that of the input image. This is needed so that the "current image" and the
# "previous mask" share the same timestamp, and as a result can be synchronized
# and combined in the subsequent calculator. Note that upon the arrival of the
# very first input frame, an empty packet is sent out to jump start the feedback
# loop.
# Caches a mask fed back from the previous round of hair segmentation, and upon
# the arrival of the next input image sends out the cached mask with the
# timestamp replaced by that of the input image, essentially generating a packet
# that carries the previous mask. Note that upon the arrival of the very first
# input image, an empty packet is sent out to jump start the feedback loop.
node {
calculator: "PreviousLoopbackCalculator"
input_stream: "MAIN:throttled_input_video"
@@ -114,9 +103,9 @@ node {
# Converts the transformed input image on GPU into an image tensor stored in
# tflite::gpu::GlBuffer. The zero_center option is set to false to normalize the
# pixel values to [0.f, 1.f] as opposed to [-1.f, 1.f].
# With the max_num_channels option set to 4, all 4 RGBA channels are contained
# in the image tensor.
# pixel values to [0.f, 1.f] as opposed to [-1.f, 1.f]. With the
# max_num_channels option set to 4, all 4 RGBA channels are contained in the
# image tensor.
node {
calculator: "TfLiteConverterCalculator"
input_stream: "IMAGE_GPU:mask_embedded_input_video"
@@ -147,7 +136,7 @@ node {
node {
calculator: "TfLiteInferenceCalculator"
input_stream: "TENSORS_GPU:image_tensor"
output_stream: "TENSORS:segmentation_tensor"
output_stream: "TENSORS_GPU:segmentation_tensor"
input_side_packet: "CUSTOM_OP_RESOLVER:op_resolver"
node_options: {
[type.googleapis.com/mediapipe.TfLiteInferenceCalculatorOptions] {
@@ -157,23 +146,15 @@ node {
}
}
# The next step (tensors to segmentation) is not yet supported on iOS GPU.
# Convert the previous segmentation mask to CPU for processing.
node: {
calculator: "GpuBufferToImageFrameCalculator"
input_stream: "previous_hair_mask"
output_stream: "previous_hair_mask_cpu"
}
# Decodes the segmentation tensor generated by the TensorFlow Lite model into a
# mask of values in [0.f, 1.f], stored in the R channel of a CPU buffer. It also
# mask of values in [0.f, 1.f], stored in the R channel of a GPU buffer. It also
# takes the mask generated previously as another input to improve the temporal
# consistency.
node {
calculator: "TfLiteTensorsToSegmentationCalculator"
input_stream: "TENSORS:segmentation_tensor"
input_stream: "PREV_MASK:previous_hair_mask_cpu"
output_stream: "MASK:hair_mask_cpu"
input_stream: "TENSORS_GPU:segmentation_tensor"
input_stream: "PREV_MASK_GPU:previous_hair_mask"
output_stream: "MASK_GPU:hair_mask"
node_options: {
[type.googleapis.com/mediapipe.TfLiteTensorsToSegmentationCalculatorOptions] {
tensor_width: 512
@@ -185,13 +166,6 @@ node {
}
}
# Send the current segmentation mask to GPU for the last step, blending.
node: {
calculator: "ImageFrameToGpuBufferCalculator"
input_stream: "hair_mask_cpu"
output_stream: "hair_mask"
}
# Colors the hair segmentation with the color specified in the option.
node {
calculator: "RecolorCalculator"