OpenVX Samples

Khronos OpenVX™ is an open, royalty-free standard for cross-platform acceleration of computer vision applications. OpenVX enables performance and power-optimized computer vision processing, especially important in embedded and real-time use cases such as face, body, and gesture tracking, smart video surveillance, advanced driver assistance systems (ADAS), object and scene reconstruction, augmented reality, visual inspection, robotics and more.

In this project, we provide OpenVX sample applications to use with any conformant implementation of OpenVX.

• VX Bubble Pop Sample
• VX Canny Edge Detector Sample
• VX Skin Tone Detector Sample

VX Bubble Pop Sample

In this sample we will create an OpenVX graph to run VX Bubble Pop on a live camera. This sample application uses OpenCV to decode input image, draw bubbles/donuts and display the output.

Prerequisites

• Conformant OpenVX Implementation

• OpenCV

• Camera

Steps to run the Bubble Pop sample

• Step - 1: Build and install Conformant OpenVX Implementation. In this example we will use the OpenVX Sample Implementation available on GitHub

Build OpenVX on Linux

* Git Clone project with a recursive flag to get submodules

      git clone --recursive https://github.com/KhronosGroup/OpenVX-sample-impl.git

* Use Build.py script

      cd OpenVX-sample-impl/
      python Build.py --os=Linux --arch=64 --conf=Debug --conf_vision --enh_vision --conf_nn

• Step - 2: Export OpenVX Directory Path

export OPENVX_DIR=$(pwd)/install/Linux/x64/Debug

• Step - 3: Clone the OpenVX Samples project and build the bubble pop application

cd ~/ && mkdir OpenVXSample-pop
cd OpenVXSample-pop/
git clone https://github.com/kiritigowda/openvx-samples.git

• Step - 4: CMake and Build the pop application

mkdir pop-build && cd pop-build
cmake -DOPENVX_INCLUDES=$OPENVX_DIR/include -DOPENVX_LIBRARIES=$OPENVX_DIR/bin/libopenvx.so ../openvx-samples/bubble-pop/
make

• Step - 5: Run VX Pop application

  • Bubbles

  ./vxPop --bubble
  

  • Donuts

  ./vxPop --donut
  

Canny Edge Detector Sample

In this sample we will create an OpenVX graph to run canny edge detection on an image or a live camera. This sample application uses OpenCV to decode input image and display the output.

Prerequisites

• Conformant OpenVX Implementation

• OpenCV

Steps to run the canny sample

• Step - 1: Build and install Conformant OpenVX Implementation. In this example we will use the OpenVX Sample Implementation available on GitHub

Build OpenVX on Linux

* Git Clone project with a recursive flag to get submodules

      git clone --recursive https://github.com/KhronosGroup/OpenVX-sample-impl.git

* Use Build.py script

      cd OpenVX-sample-impl/
      python Build.py --os=Linux --arch=64 --conf=Debug --conf_vision --enh_vision --conf_nn

• Step - 2: Export OpenVX Directory Path

export OPENVX_DIR=$(pwd)/install/Linux/x64/Debug

• Step - 3: Clone the OpenVX Samples project and build the canny application

cd ~/ && mkdir OpenVXSample-canny
cd OpenVXSample-canny/
git clone https://github.com/kiritigowda/openvx-samples.git

• Step - 4: CMake and Build the canny application

mkdir canny-build && cd canny-build
cmake -DOPENVX_INCLUDES=$OPENVX_DIR/include -DOPENVX_LIBRARIES=$OPENVX_DIR/bin/libopenvx.so ../openvx-samples/canny-edge-detector/
make

• Step - 5: Run Canny application

  • Live

  ./cannyEdgeDetector --live
  

  • Image

  ./cannyEdgeDetector --image ../openvx-samples/images/face.png
  

Skin Tone Detector Sample

In this sample we will create an OpenVX graph to run skintone detection on an image or a live camera. This sample application uses OpenCV to decode input image and display the output.

Prerequisites

• Conformant OpenVX Implementation

• OpenCV

Steps to run the skin tone sample

• Step - 1: Build and install Conformant OpenVX Implementation. In this example we will use the OpenVX Sample Implementation available on GitHub

Build OpenVX on Linux

* Git Clone project with a recursive flag to get submodules

      git clone --recursive https://github.com/KhronosGroup/OpenVX-sample-impl.git

* Use Build.py script

      cd OpenVX-sample-impl/
      python Build.py --os=Linux --arch=64 --conf=Debug --conf_vision --enh_vision --conf_nn

• Step - 2: Export OpenVX Directory Path

export OPENVX_DIR=$(pwd)/install/Linux/x64/Debug

• Step - 3: Clone the OpenVX Samples project and build the Skin Tone application

cd ~/ && mkdir OpenVXSample-skintone
cd OpenVXSample-skintone/
git clone https://github.com/kiritigowda/openvx-samples.git

• Step - 4: CMake and Build the Skin Tone application

mkdir skintone-build && cd skintone-build
cmake -DOPENVX_INCLUDES=$OPENVX_DIR/include -DOPENVX_LIBRARIES=$OPENVX_DIR/bin/libopenvx.so ../openvx-samples/skin-tone-detector/
make

• Step - 5: Run Skin Tone Detector application

  • Live

  ./skinToneDetector --live
  

  • Image

  ./skinToneDetector --image ../openvx-samples/images/face.png
  

Contribution

The samples VX Bubble Pop, VX Canny Edge Detector, & VX Skin Tone Detector are contributed by AMD from their MIVisionX Toolkit. We welcome contributions to this project from all developers. Please open a pull request with details of your sample application to be accepted into this project.