Mediapipe

* MediaPipe's Abseil dependency is now pointing to [abseil's LTS_2020_02_25 release](https://github.com/abseil/abseil-cpp/releases/tag/20200225).
* MediaPipe is now switched to [absl::Status](https://github.com/abseil/abseil-cpp/tree/master/absl/status).
* MediaPipe's TensorFlow version is now pointing to [commit 77e9ffb9b2bfb1a4f7056e62d84039626923e328](https://github.com/tensorflow/tensorflow/commit/77e9ffb9b2bfb1a4f7056e62d84039626923e328), which is submitted on Feb 12, 2020.
* The range of Bazel version that MediaPipe supports is now narrowed. We support Bazel from v1.0.0 to v1.2.1.

* improved android camera support (camera properties including).
* subgraph node and stream naming improvement
* bug fix for data race in SimulationClockExecutor
* concurrency macros update for ABSL (GUARDED_BY --> ABSL_GUARDED_BY)
* image_cropping_calculator improvement.
* new utility calculator, ConstantSidePacketCalculato
* Web demos
* [Edge Detection](https://viz.mediapipe.dev/runner/demos/edge_detection/edge_detection.html)
* [Face Detection](https://viz.mediapipe.dev/runner/demos/face_detection/face_detection.html)
* [Hands Tracking](https://viz.mediapipe.dev/runner/demos/hand_tracking/hand_tracking.html)
* [Hair Segmentation](https://viz.mediapipe.dev/runner/demos/hair_segmentation/hair_segmentation.html)
* Documentation
* Corel example improvement
* [YouTube channel](https://www.youtube.com/channel/UCObqmpuSMx-usADtL_qdMAw)
* [MediaPipe on the Web - Google Dev Blog Post](https://developers.googleblog.com/2020/01/mediapipe-on-web.html)

* [MediaPipe AutoFlip desktop example](https://github.com/google/mediapipe/tree/master/mediapipe/examples/desktop/autoflip)
* Bazel maximum version check (version 1.2.1)
* [Added way to extract detection protos from object detection android mobile GPU example](https://github.com/google/mediapipe/commit/ae6be10afe59a6a99d8a68007784706ac98720dd#diff-e2d9229283509a972282c1d03f41ec3d)
* Bug fixes
* previous_loopback_calculator
* split_vector_calculator
* scale_image_calculator
* Documentation
* [MediaPipe AutoFlip documentation](https://github.com/google/mediapipe/blob/master/mediapipe/docs/autoflip.md)
* Fixes for [framework concept](https://github.com/google/mediapipe/commit/ae6be10afe59a6a99d8a68007784706ac98720dd#diff-31805dad11a7299bfc5fa68e64a3611a), [coral face](https://github.com/google/mediapipe/commit/ae6be10afe59a6a99d8a68007784706ac98720dd#diff-44a9f2c7d5adb5e036037bc747a4365a) and [object detection](https://github.com/google/mediapipe/commit/ae6be10afe59a6a99d8a68007784706ac98720dd#diff-dab963cc487732c4c0640be442119e2c) docs

* [MediaPipe Tracking library](https://github.com/google/mediapipe/tree/master/mediapipe/util/tracking)
* MediaPipe Object Detection (GPU) & Tracking example [android](https://github.com/google/mediapipe/blob/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/multihandtrackinggpu/MainActivity.java)
* Bug fixes
* (Multi)handtracking demo crashes on Mali and PowerVR GPUs [android]
* iOS fixes - better documentation (https://github.com/google/mediapipe/issues/25, https://github.com/google/mediapipe/issues/135), iOS crash (https://github.com/google/mediapipe/issues/145)
* .Documentation
* [MediaPipe Object Detection (GPU) & Tracking](https://github.com/google/mediapipe/blob/master/mediapipe/docs/object_tracking_mobile_gpu.md)
* [Google Developer MediaPipe Object Tracking blog post](https://mediapipe.page.link/objecttrackingblog)

* Multihands tracking example to get hand landmarks
* [android](https://github.com/google/mediapipe/blob/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/multihandtrackinggpu/MainActivity.java)
* [ios](https://github.com/google/mediapipe/blob/master/mediapipe/examples/ios/multihandtrackinggpu/ViewController.mm)
* Hands tracking example to get hand landmarks
* [android](https://github.com/google/mediapipe/blob/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/handtrackinggpu/MainActivity.java)
* [ios](https://github.com/google/mediapipe/blob/master/mediapipe/examples/ios/handtrackinggpu/ViewController.mm)
* [Multi-hand Tracking AAR Example](https://github.com/jiuqiant/mediapipe_multi_hands_tracking_aar_example)
* Add basic audio support to opencv video decoder and encoder calculator
* Add NNAPI inference support to TfLiteInferenceCalculator on Android.
* Bug fixes
* Switch to Abseil LTS version [issue 264](https://github.com/google/mediapipe/issues/264)
* Downgrade to use Bazel 1.1.0 on macOS
* Fix a docker setup issue [issue 276](https://github.com/google/mediapipe/issues/276)
* Access Hand landmarks in Android & iOS issues [219](https://github.com/google/mediapipe/issues/219), [230](https://github.com/google/mediapipe/issues/230), [79](https://github.com/google/mediapipe/issues/79), [64](https://github.com/google/mediapipe/issues/64), [237](https://github.com/google/mediapipe/issues/237)

* MediaPipe on Coral dev board EdgeTPU
* Dockerfile for cross compiling
* [Example including quantized models & graphs for Object detection & Face detection](https://github.com/google/mediapipe/blob/master/mediapipe/docs/examples.md#google-coral-machine-learning-acceleration-with-google-edgetpu)
* Bug fixes
* iOS Tulsi setup bug
* TFLite OpenGL ES delegate fix integration that fixes [issue 219](https://github.com/google/mediapipe/issues/219), [issue 232](https://github.com/google/mediapipe/issues/232), [issue 104](https://github.com/google/mediapipe/issues/104), [issue 87](https://github.com/google/mediapipe/issues/87), and [issue 41](https://github.com/google/mediapipe/issues/41)
* Documentation
* [Setup MediaPipe on Coral dev board Edge TPU](https://github.com/google/mediapipe/blob/master/mediapipe/examples/coral/README.md)
* [Coral Object detection example](https://github.com/google/mediapipe/blob/master/mediapipe/docs/object_detection_coral_devboard.md)
* [Coral Face detection example](https://github.com/google/mediapipe/blob/master/mediapipe/docs/face_detection_coral_devboard.md)