ermig1979/Simd

Health v2 · maintenance only

Adoption not part of the score

2265 stars · 453 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Simd Library is a free open-source C++ image processing and machine learning library with a C API and Python wrapper. Its algorithms are hand-optimized with SIMD CPU extensions including SSE, AVX, AVX-512, AMX (x86/x64), NEON, SVE/SVE2 (ARM), and HVX (Hexagon).

Use cases

• accelerate image scaling, filtering, and pixel format conversion with SIMD
• detect motion in video streams
• perform object detection with HAAR and LBP cascades
• run neural network inference on CPU with SIMD optimizations
• extract statistical information from images
• optimize C++ image pipelines for x86, ARM, and Hexagon targets

When to choose

• you need maximum CPU performance for image processing in C/C++
• you target multiple architectures (x86/x64, ARM, Hexagon) with SIMD-optimized code
• you need lightweight CPU-based object detection or neural network inference without heavy frameworks

When to avoid

• you need GPU acceleration
• you want a high-level Python-first computer vision framework
• you need a broad ecosystem of pretrained models and community tooling like OpenCV

Facets

library · maturity active

image-processing machine-learning computer-vision deep-learning image-processing computer-vision machine-learning performance windows cpp python cross-platform simd sse avx avx-512 amx neon sve hexagon-hvx haar-cascade object-detection motion-detection neural-network-inference c-api linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ermig1979/Simd")