Donde server

It's a group of different services

• image feature extraction
• feature searching & clustering
• video processing

Built on top of donde-toolkits

Features

• Openvino inference
• Full pipeline detect/landmarks/align/extract
• Conan to manage dependencies
• Protobuf proto & GRPC server
• GRPC & GRPC gateway integrated

Dependencies

For the big ones: openvino/opencv/ffmpeg, other small depencencies are managed by conan/conanfile.txt.

Models

name	type	desc	urldesc
face-detection-adas-0001	detect	detect face	https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/intel/face-detection-adas-0001/README.md
facial-landmarks-35-adas-0002	landmarks	70 points landmarks	https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/intel/facial-landmarks-35-adas-0002/README.md
Sphereface.xml	feature	extract face feature	https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/Sphereface/README.md

Project layout

• proto contains protobuf definitions for common api, grpc services definitions. servers are typically implementations of these grpc services.

• api proto files are compiled to api directory.

• src has source codes to build various standalone server binaries.

• tests contains tests, they test the FeatureLibraries library. tests has dependency to library.

• contrib contains json samples, static assets etc.

Usage

Build and run

Install dependencies

macOS

brew install openvino
brew install ffmpeg
brew install opencv

other platform openvino installation guide, please refers to https://www.intel.com/content/www/us/en/developer/tools/openvino-toolkit/download.html

find_package(OpenVINO REQUIRED COMPONENTS Runtime)

Build conan dependency packages

mkdir -p generated
conan install --build=missing  -of generated ./conan

Build

mkdir -p build && cd build
cmake -DCMAKE_TOOLCHAIN_FILE=../generated/conan_toolchain.cmake -DCMAKE_BUILD_TYPE=Debug ..
cmake --build .

./build/feature_extract --config_path contrib/face_extract_server.json

Build proto

cd proto

#build protobuf definitions of grpc and service
./build_proto.sh

#build grpc - http gateway, for http api[optional]
./build_gateway.sh

Xcode

mkdir -p build && cd build
cmake -G Xcode -DCMAKE_TOOLCHAIN_FILE=../generated/conan_toolchain.cmake ..

then use xcode to open FaceRecognitionServer.xcodeproj, try build and run.

FAQ

Q: CMake generate failed with message xcode-select: error: tool 'xcodebuild' requires Xcode, but active developer directory '/Library/Developer/CommandLineTools' is a command line tools instance

A: Run sudo xcode-select -s /Applications/Xcode.app/Contents/Developer, see also https://github.com/nodejs/node-gyp/issues/569.

Q: Build failed with message "Protobuf requires at least C++11."

A: Navigate to TARGETS -> API, find Apple Clang - Language - C++, change C++ Language Dialect to C++17[-std=c++17]

Q: Run failed with message dyld[4593]: Library not loaded: @rpath/libopenvinod.dylib

A:

It's because we need source /opt/intel/openvino_2022/setupvars.sh, but xcode doesn't know that. Edit FaceRecognitionServer scheme, add environment variable DYLD_LIBRARY_PATH, value /opt/intel/openvino_2022/runtime/3rdparty/tbb/lib::/opt/intel/openvino_2022/runtime/lib/intel64/Release:/opt/intel/openvino_2022/runtime/lib/intel64/Debug

A2:

for i in `ls /opt/intel/openvino_2022/runtime/lib/intel64/Debug/`; do sudo ln -s /opt/intel/openvino_2022/runtime/lib/intel64/Debug/$i /usr/local/lib; done
for i in `ls /opt/intel/openvino_2022/runtime/3rdparty/tbb/lib/`; do sudo ln -s /opt/intel/openvino_2022.1.0.643/runtime/3rdparty/tbb/lib/$i /usr/local/lib; done

Mac M1

README mac m1