Simd
C++ image processing and machine learning library with using of SIMD: SSE, AVX, AVX-512, AMX for x86/x64, NEON, SVE for ARM, HVX for Hexagon
What is Simd?
C++ image processing and machine learning library with using of SIMD: SSE, AVX, AVX-512, AMX for x86/x64, NEON, SVE for ARM, HVX for Hexagon
Overview of Simd
What it is
Simd is a free open source C and C++ library for image processing and machine learning. It provides high performance algorithms for tasks such as pixel format conversion, image scaling and filtration, extraction of statistic information from images, motion detection, and neural network operations. The library is optimized using SIMD CPU extensions including SSE, AVX, AVX-512, and AMX for x86/x64, and NEON, SVE, SVE2 for ARM. It has a C API and includes C++ classes and functions to facilitate access to the C API. The library supports dynamic and static linking, 32-bit and 64-bit Windows and Linux, MSVS, G++ and Clang compilers, and MSVS project and CMake build systems.
Key features
- Provides many high performance algorithms for image processing: pixel format conversion, image scaling and filtration, extraction of statistic information, motion detection, and neural network.
- Optimized with SIMD CPU extensions: SSE, AVX, AVX-512, AMX for x86/x64; NEON, SVE, SVE2 for ARM.
- Has a C API and C++ classes/functions to facilitate access to the C API.
- Supports dynamic and static linking, 32-bit and 64-bit Windows and Linux, MSVS, G++ and Clang compilers, MSVS project and CMake build systems.
- Includes a test framework for correctness and performance testing.
- Provides a Python wrapper (SimdPy) and supports mutual conversion with OpenCV types when the macro SIMD_OPENCV_ENABLE is defined.
- Can be installed via the vcpkg dependency manager.
Use cases
Simd is suitable for developers who need high performance image processing and machine learning algorithms in C or C++ applications. It can be used in computer vision pipelines, real-time video analysis, image format conversion, motion detection systems, and neural network inference on CPU. The library is designed for use on x86/x64 and ARM platforms, including cross-compilation for ARM targets.
Getting started
To build the library on Linux using CMake and the native compiler:shmkdir buildcd buildcmake ../prj/cmake -DSIMD_TOOLCHAIN='' -DSIMD_TARGET=''makeTo build on Windows with Microsoft Visual Studio 2022, use the project files in simd/prj/vs2022/. For installation via vcpkg:shgit clone https://github.com/Microsoft/vcpkg.gitcd vcpkg./bootstrap-vcpkg.sh./vcpkg integrate install./vcpkg install simd
License and pricing
The library is licensed under the MIT license. It is open source and free to use.
Based on the project's public README and repository metadata.