Latest Release

MIVisionX toolkit is a set of comprehensive computer vision and machine intelligence libraries, utilities, and applications bundled into a single toolkit. AMD MIVisionX delivers highly optimized conformant open-source implementation of the Khronos OpenVX™ and OpenVX™ Extensions along with Convolution Neural Net Model Compiler & Optimizer supporting ONNX, and Khronos NNEF™ exchange formats. The toolkit allows for rapid prototyping and deployment of optimized computer vision and machine learning inference workloads on a wide range of computer hardware, including small embedded x86 CPUs, APUs, discrete GPUs, and heterogeneous servers.

Latest Release#

Table of Contents#

Documentation#

Run the steps below to build documentation locally.

cd docs
pip3 install -r .sphinx/requirements.txt
python3 -m sphinx -T -E -b html -d _build/doctrees -D language=en . _build/html

AMD OpenVX™#

AMD OpenVX™ is a highly optimized conformant open source implementation of the Khronos OpenVX™ 1.3 computer vision specification. It allows for rapid prototyping as well as fast execution on a wide range of computer hardware, including small embedded x86 CPUs and large workstation discrete GPUs.

Khronos OpenVX™ 1.0.1 conformant implementation is available in MIVisionX Lite

AMD OpenVX™ Extensions#

The OpenVX framework provides a mechanism to add new vision functionality to OpenVX by vendors. This project has below mentioned OpenVX modules and utilities to extend amd_openvx, which contains the AMD OpenVX™ Core Engine.

amd_loomsl: AMD Radeon Loom stitching library for live 360 degree video applications
amd_media: vx_amd_media is an OpenVX AMD media extension module for encode and decode
amd_migraphx: amd_migraphx extension integrates the AMD’s MIGraphx into an OpenVX graph. This extension allows developers to combine the vision funcions in OpenVX with the MIGraphX and build an end-to-end application for inference.
amd_nn: OpenVX neural network module
amd_opencv: OpenVX module that implements a mechanism to access OpenCV functionality as OpenVX kernels
amd_rpp: OpenVX extension providing an interface to some of the RPP’s (ROCm Performance Primitives) functions. This extension is used to enable rocAL to perform image augmentation.
amd_winml: WinML extension will allow developers to import a pre-trained ONNX model into an OpenVX graph and add hundreds of different pre & post processing vision / generic / user-defined functions, available in OpenVX and OpenCV interop, to the input and output of the neural net model. This will allow developers to build an end to end application for inference.

Applications#

MIVisionX has several applications built on top of OpenVX modules, it uses AMD optimized libraries to build applications that can be used to prototype or use as a model to develop products.

Bubble Pop: This sample application creates bubbles and donuts to pop using OpenVX & OpenCV functionality.
Cloud Inference Application: This sample application does inference using a client-server system.
Digit Test: This sample application is used to recognize hand written digits.
Image Augmentation: This sample application demonstrates the basic usage of rocAL’s C API to load JPEG images from the disk and modify them in different possible ways and displays the output images.
MIVisionX Inference Analyzer: This sample application uses pre-trained ONNX / NNEF / Caffe models to analyze and summarize images.
MIVisionX OpenVX Classification: This sample application shows how to run supported pre-trained caffe models with MIVisionX RunTime.
MIVisionX Validation Tool: This sample application uses pre-trained ONNX / NNEF / Caffe models to analyze, summarize and validate models.
MIVisionX WinML Classification: This sample application shows how to run supported ONNX models with MIVisionX RunTime on Windows.
MIVisionX WinML YoloV2: This sample application shows how to run tiny yolov2(20 classes) with MIVisionX RunTime on Windows.
Optical Flow: This sample application creates an OpenVX graph to run Optical Flow on a video/live.
External Applications

Neural Net Model Compiler & Optimizer#

Neural Net Model Compiler & Optimizer converts pre-trained neural net models to MIVisionX runtime code for optimized inference.

rocAL#

The ROCm Augmentation Library - rocAL is designed to efficiently decode and process images and videos from a variety of storage formats and modify them through a processing graph programmable by the user.

Toolkit#

MIVisionX Toolkit, is a comprehensive set of helpful tools for neural net creation, development, training, and deployment. The Toolkit provides you with helpful tools to design, develop, quantize, prune, retrain, and infer your neural network work in any framework. The Toolkit is designed to help you deploy your work to any AMD or 3rd party hardware, from embedded to servers.

MIVisionX provides you with tools for accomplishing your tasks throughout the whole neural net life-cycle, from creating a model to deploying them for your target platforms.

Utilities#

inference_generator: generate inference library from pre-trained CAFFE models
loom_shell: an interpreter to prototype 360 degree video stitching applications using a script
RunVX: command-line utility to execute OpenVX graph described in GDF text file
RunCL: command-line utility to build, execute, and debug OpenCL programs

Prerequisites#

Hardware#

CPU: AMD64
GPU: AMD Radeon™ Graphics [optional]
APU: AMD Radeon™ Mobile/Embedded [optional]

Note: Some modules in MIVisionX can be built for CPU ONLY. To take advantage of Advanced Features And Modules we recommend using AMD GPUs or AMD APUs.

Operating System#

Windows#

Windows 10 / 11
Windows SDK
Visual Studio 2019 or later
Install the latest AMD drivers
Install OpenCL SDK
Install OpenCV 4.6.0
- Set OpenCV_DIR environment variable to OpenCV/build folder
- Add %OpenCV_DIR%\x64\vc14\bin or %OpenCV_DIR%\x64\vc15\bin to your PATH

macOS#

macOS - Ventura 13.4
Install Homebrew
Install CMake
Install OpenCV 3/4

Note: macOS build instructions

Linux#

Linux distribution
- Ubuntu - 20.04 / 22.04
- CentOS - 7 / 8
- RedHat - 8 / 9
- SLES - 15-SP4
Install ROCm
CMake 3.0 or later
ROCm MIOpen for Neural Net Extensions (vx_nn)
Qt Creator for Cloud Inference Client
Protobuf for inference generator & model compiler
- install libprotobuf-dev and protobuf-compiler needed for vx_nn
OpenCV 4.6.0
- Set OpenCV_DIR environment variable to OpenCV/build folder
FFMPEG n4.4.2
- FFMPEG is required for amd_media & mv_deploy modules
rocAL Prerequisites

Prerequisites setup script for Linux#

For the convenience of the developer, we provide the setup script MIVisionX-setup.py which will install all the dependencies required by this project.

NOTE: This script only needs to be executed once.

Prerequisites for running the script#

Linux distribution
- Ubuntu - 20.04 / 22.04
- CentOS - 7 / 8
- RedHat - 8 / 9
- SLES - 15-SP4
ROCm supported hardware

ROCm

usage:

python MIVisionX-setup.py --directory [setup directory - optional (default:~/)]
                          --opencv    [OpenCV Version - optional (default:4.6.0)]
                          --protobuf  [ProtoBuf Version - optional (default:3.12.4)]
                          --rpp       [RPP Version - optional (default:1.0.0)]
                          --pybind11  [PyBind11 Version - optional (default:v2.10.4)]
                          --ffmpeg    [FFMPEG V4.4.2 Installation - optional (default:ON) [options:ON/OFF]]
                          --rocal     [MIVisionX rocAL Dependency Install - optional (default:ON) [options:ON/OFF]]
                          --neural_net[MIVisionX Neural Net Dependency Install - optional (default:ON) [options:ON/OFF]]
                          --inference [MIVisionX Neural Net Inference Dependency Install - optional (default:ON) [options:ON/OFF]]
                          --developer [Setup Developer Options - optional (default:ON) [options:ON/OFF]]
                          --reinstall [Remove previous setup and reinstall (default:OFF)[options:ON/OFF]]
                          --backend   [MIVisionX Dependency Backend - optional (default:HIP) [options:HIP/OCL/CPU]]
                          --rocm_path [ROCm Installation Path - optional (default:/opt/rocm) - ROCm Installation Required]

Note:

ROCm upgrade with sudo apt upgrade requires the setup script rerun.
use X Window / X11 for remote GUI app control

Build & Install MIVisionX#

Building on Windows#

Using `Visual Studio`#

Install Windows Prerequisites
Use MIVisionX.sln to build for x64 platform

NOTE: vx_nn is not supported on Windows in this release

Building on macOS#

macOS build instructions

Building on Linux#

Using `apt-get` or `yum`#

ROCm supported hardware
Install ROCm
On Ubuntu
```
sudo apt-get install mivisionx
```
On CentOS
```
sudo yum install mivisionx
```
Note:
- vx_winml is not supported on Linux
- source code will not available with apt-get / yum install
- the installer will copy
  - Executables into /opt/rocm/bin
  - Libraries into /opt/rocm/lib
  - OpenVX and module header files into /opt/rocm/include/mivisionx
  - Model compiler, & toolkit folders into /opt/rocm/libexec/mivisionx
  - Apps, & samples folder into /opt/rocm/share/mivisionx
  - Docs folder into /opt/rocm/share/doc/mivisionx
- Package (.deb & .rpm) install requires OpenCV v3+ to execute AMD OpenCV extensions

Using MIVisionX-setup.py#

Install ROCm
Use the below commands to set up and build MIVisionX
```
git clone https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX.git
cd MIVisionX
```
Note: MIVisionX has support for two GPU backends: OPENCL and HIP:
- Instructions for building MIVisionX with the HIP GPU backend (i.e., default GPU backend):
  - run the setup script to install all the dependencies required by the HIP GPU backend:
```
python MIVisionX-setup.py
```
  - run the below commands to build MIVisionX with the HIP GPU backend:
```
mkdir build-hip
cd build-hip
cmake ../
make -j8
sudo cmake --build . --target PyPackageInstall
sudo make install
```
  Note:
  - PyPackageInstall used for rocal_pybind installation
  - rocal_pybind not supported on windows.
  - sudo required for pybind installation
- Instructions for building MIVisionX with OPENCL GPU backend:
  - run the setup script to install all the dependencies required by the OPENCL GPU backend:
```
python MIVisionX-setup.py --reinstall ON --backend OCL
```
  - run the below commands to build MIVisionX with the OPENCL GPU backend:
```
mkdir build-ocl
cd build-ocl
cmake -DBACKEND=OPENCL ../
make -j8
sudo make install
```
Note:
- MIVisionX cannot be installed for both GPU backends in the same default folder (i.e., /opt/rocm/) if an app interested in installing MIVisionX with both GPU backends, then add -DCMAKE_INSTALL_PREFIX in the cmake commands to install MIVisionX with OPENCL and HIP backends into two separate custom folders.
- vx_winml is not supported on Linux
- PyPackageInstall used for rocal_pybind installation - supported only with HIP Backend
- rocal_pybind not supported on windows
- sudo required for pybind installation

Verify the Installation#

Verifying on Linux / macOS#

The installer will copy
- Executables into /opt/rocm/bin
- Libraries into /opt/rocm/lib
- OpenVX and OpenVX module header files into /opt/rocm/include/mivisionx
- Apps, & Samples folder into /opt/rocm/share/mivisionx
- Documents folder into /opt/rocm/share/doc/mivisionx
- Model Compiler, and Toolkit folder into /opt/rocm/libexec/mivisionx
Run the below sample to verify the installation

Canny Edge Detection
```
export PATH=$PATH:/opt/rocm/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/rocm/lib
runvx /opt/rocm/share/mivisionx/samples/gdf/canny.gdf
```
Note: More samples are available here

Note: For macOS use export DYLD_LIBRARY_PATH=$DYLD_LIBRARY_PATH:/opt/rocm/lib

Verifying on Windows#

MIVisionX.sln builds the libraries & executables in the folder MIVisionX/x64

Use RunVX to test the build

./runvx.exe PATH_TO/MIVisionX/samples/gdf/skintonedetect.gdf

Docker#

MIVisionX provides developers with docker images for Ubuntu 20.04 and CentOS 7 / 8. Using docker images developers can quickly prototype and build applications without having to be locked into a single system setup or lose valuable time figuring out the dependencies of the underlying software.

Docker files to build MIVisionX containers are available

MIVisionX Docker#

Docker Workflow Sample on Ubuntu `20.04`#

Sample Prerequisites#

Ubuntu 20.04/22.04
rocm supported hardware

Sample Workflow#

Step 1 - Install rocm-dkms

sudo apt update -y
sudo apt dist-upgrade -y
sudo apt install libnuma-dev wget
sudo reboot

wget https://repo.radeon.com/amdgpu-install/21.50/ubuntu/focal/amdgpu-install_21.50.50000-1_all.deb
sudo apt-get install -y ./amdgpu-install_21.50.50000-1_all.deb
sudo apt-get update -y
sudo amdgpu-install -y --usecase=rocm
sudo reboot

Step 2 - Setup Docker

sudo apt-get install curl
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt-get update
apt-cache policy docker-ce
sudo apt-get install -y docker-ce
sudo systemctl status docker

Step 3 - Get Docker Image

sudo docker pull mivisionx/ubuntu-20.04

Step 4 - Run the docker image

sudo docker run -it --device=/dev/kfd --device=/dev/dri --cap-add=SYS_RAWIO --device=/dev/mem --group-add video --network host mivisionx/ubuntu-20.04:latest

Note:

Map host directory on the docker image

map the localhost directory to be accessed on the docker image.
use -v option with docker run command: -v {LOCAL_HOST_DIRECTORY_PATH}:{DOCKER_DIRECTORY_PATH}
usage:

sudo docker run -it -v /home/:/root/hostDrive/ --device=/dev/kfd --device=/dev/dri --cap-add=SYS_RAWIO --device=/dev/mem --group-add video --network host mivisionx/ubuntu-20.04:latest

Display option with docker

Using host display

xhost +local:root
sudo docker run -it --device=/dev/kfd --device=/dev/dri --cap-add=SYS_RAWIO --device=/dev/mem --group-add video --network host --env DISPLAY=unix$DISPLAY --privileged --volume $XAUTH:/root/.Xauthority --volume /tmp/.X11-unix/:/tmp/.X11-unix mivisionx/ubuntu-20.04:latest

Test display with MIVisionX sample

export PATH=$PATH:/opt/rocm/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/rocm/lib
runvx /opt/rocm/share/mivisionx/samples/gdf/canny.gdf

Technical Support#

Please email mivisionx.support@amd.com for questions, and feedback on MIVisionX.

Please submit your feature requests, and bug reports on the GitHub issues page.

Release Notes#

Latest Release Version#

Changelog#

Review all notable changes with the latest release

Tested configurations#

Windows 10 / 11
Linux distribution
- Ubuntu - 20.04 / 22.04
- CentOS - 7 / 8
- RHEL - 8 / 9
- SLES - 15-SP4
ROCm: rocm-core - 5.4.3.50403-121
miopen-hip - 2.19.0.50403-121
miopen-opencl - 2.18.0.50300-63
migraphx - 2.4.0.50403-121
Protobuf - V3.12.4
OpenCV - 4.6.0
RPP - 1.2.0
FFMPEG - n4.4.2
Dependencies for all the above packages
MIVisionX Setup Script - V2.5.4

Known issues#

OpenCV 4.X support for some apps missing

MIVisionX Package install requires manual prerequisites installation: some prerequisites for MIVisionX are build from source and do not have equivalent packages in linux. The user has to manually install these prerequisites. Instructions for these is given below.

Boost - Version 1.72.0

wget https://boostorg.jfrog.io/artifactory/main/release/1.72.0/source/boost_1_72_0.tar.bz2
tar xjvf boost_1_72_0.tar.bz2
cd ./boost_1_72_0/
./bootstrap.sh --prefix=/usr/local --with-python=python3
./b2 stage -j16 threading=multi link=shared cxxflags="-std=c++11"
sudo ./b2 install threading=multi link=shared --with-system --with-filesystem
./b2 stage -j16 threading=multi link=static cxxflags="-std=c++11 -fpic" cflags="-fpic"
sudo ./b2 install threading=multi link=static --with-system --with-filesystem

OpenCV - Version 4.6.0

wget https://github.com/opencv/opencv/archive/4.6.0.zip
unzip 4.6.0.zip
mkdir -p build/OpenCV
cd build/OpenCV
cmake -D WITH_GTK=ON -D WITH_JPEG=ON -D BUILD_JPEG=ON -D WITH_OPENCL=OFF -D WITH_OPENCLAMDFFT=OFF -D WITH_OPENCLAMDBLAS=OFF -D WITH_VA_INTEL=OFF -D WITH_OPENCL_SVM=OFF  -D CMAKE_INSTALL_PREFIX=/usr/local ../../opencv-4.6.0
make -j8
make install
ldconfig

FFMPEG - Version n4.4.2

Turbo JPEG installation - Version 2.0.6.1

git clone -b 2.0.6.1 https://github.com/rrawther/libjpeg-turbo.git
cd libjpeg-turbo && mkdir build && cd build
cmake -DCMAKE_INSTALL_PREFIX=/usr \
  -DCMAKE_BUILD_TYPE=RELEASE  \
  -DENABLE_STATIC=FALSE       \
  -DCMAKE_INSTALL_DOCDIR=/usr/share/doc/libjpeg-turbo-2.0.3 \
  -DCMAKE_INSTALL_DEFAULT_LIBDIR=lib  \
  ..
make -j$nproc && sudo make install

MIVisionX Dependency Map#

HIP Backend#

Docker Image: sudo docker build -f docker/ubuntu20/{DOCKER_LEVEL_FILE_NAME}.dockerfile -t {mivisionx-level-NUMBER} .

new component added to the level
existing component from the previous level

Build Level	MIVisionX Dependencies	Modules	Libraries and Executables
`Level_1`	cmake gcc g++	amd_openvx utilities	`libopenvx.so` - OpenVX™ Lib - CPU `libvxu.so` - OpenVX™ immediate node Lib - CPU `runvx` - OpenVX™ Graph Executor - CPU with Display OFF
`Level_2`	ROCm HIP +Level 1	amd_openvx amd_openvx_extensions utilities	`libopenvx.so` - OpenVX™ Lib - CPU/GPU `libvxu.so` - OpenVX™ immediate node Lib - CPU/GPU `runvx` - OpenVX™ Graph Executor - Display OFF
`Level_3`	OpenCV FFMPEG +Level 2	amd_openvx amd_openvx_extensions utilities	`libopenvx.so` - OpenVX™ Lib `libvxu.so` - OpenVX™ immediate node Lib `libvx_amd_media.so` - OpenVX™ Media Extension `libvx_opencv.so` - OpenVX™ OpenCV InterOp Extension `mv_compile` - Neural Net Model Compile `runvx` - OpenVX™ Graph Executor - Display ON
`Level_4`	MIOpenGEMM MIOpen ProtoBuf +Level 3	amd_openvx amd_openvx_extensions apps utilities	`libopenvx.so` - OpenVX™ Lib `libvxu.so` - OpenVX™ immediate node Lib `libvx_amd_media.so` - OpenVX™ Media Extension `libvx_opencv.so` - OpenVX™ OpenCV InterOp Extension `mv_compile` - Neural Net Model Compile `runvx` - OpenVX™ Graph Executor - Display ON `libvx_nn.so` - OpenVX™ Neural Net Extension
`Level_5`	AMD_RPP rocAL deps +Level 4	amd_openvx amd_openvx_extensions apps rocAL utilities	`libopenvx.so` - OpenVX™ Lib `libvxu.so` - OpenVX™ immediate node Lib `libvx_amd_media.so` - OpenVX™ Media Extension `libvx_opencv.so` - OpenVX™ OpenCV InterOp Extension `mv_compile` - Neural Net Model Compile `runvx` - OpenVX™ Graph Executor - Display ON `libvx_nn.so` - OpenVX™ Neural Net Extension `libvx_rpp.so` - OpenVX™ RPP Extension `librocal.so` - Radeon Augmentation Library `rocal_pybind.so` - rocAL Pybind Lib

OpenCL Backend#