zuoqing1988/ZQCNN
A C++ CNN inference framework supporting Windows, Linux, and ARM platforms with SIMD optimizations and face detection demos.

ZQCNN is a lightweight CNN inference runtime written in C/C++ that supports cross-platform deployment including Windows, Linux, and ARM-Linux. It provides optimized convolution implementations using SIMD instructions (SSE/AVX for x86, NEON for ARM) and integrates with BLAS libraries like Intel MKL and OpenBLAS. The framework includes demo applications for face detection and recognition tasks, and is designed to work with models converted from TensorFlow and trained via PyTorch pipelines.
Frequently asked
- What is zuoqing1988/ZQCNN?
- A C++ CNN inference framework supporting Windows, Linux, and ARM platforms with SIMD optimizations and face detection demos.
- Is ZQCNN open source?
- Yes — zuoqing1988/ZQCNN is open source, released under the MIT license.
- What language is ZQCNN written in?
- zuoqing1988/ZQCNN is primarily written in C.
- How popular is ZQCNN?
- zuoqing1988/ZQCNN has 2.2k stars on GitHub.
- Where can I find ZQCNN?
- zuoqing1988/ZQCNN is on GitHub at https://github.com/zuoqing1988/ZQCNN.