ncnn

技术交流 QQ 群：637093648 (超多大佬) 答案：卷卷卷卷卷 （已满）

Pocky QQ 群（MLIR YES!）: 677104663(超多大佬) 答案：multi-level intermediate representation

Telegram Group https://t.me/ncnnyes

Discord Channel https://discord.gg/YRsxgmF

Current building status matrix

System	CPU (32bit)	CPU (64bit)	GPU (32bit)	GPU (64bit)
Linux (GCC)			—
Linux (Clang)			—
Linux (ARM)			—	—
Linux (MIPS)			—	—
Linux (RISC-V)	—		—	—
Linux (LoongArch)	—		—	—
Windows			—
Windows (ARM)			—	—
macOS	—		—
macOS (ARM)	—		—
Android
Android-x86
iOS			—
iOS Simulator			—
WebAssembly		—	—	—
RISC-V GCC/Newlib			—	—

Support most commonly used CNN network

支持大部分常用的 CNN 网络

• Classical CNN: VGG AlexNet GoogleNet Inception ...
• Practical CNN: ResNet DenseNet SENet FPN ...
• Light-weight CNN: SqueezeNet MobileNetV1 MobileNetV2/V3 ShuffleNetV1 ShuffleNetV2 MNasNet ...
• Face Detection: MTCNN RetinaFace scrfd ...
• Detection: VGG-SSD MobileNet-SSD SqueezeNet-SSD MobileNetV2-SSDLite MobileNetV3-SSDLite ...
• Detection: Faster-RCNN R-FCN ...
• Detection: YOLOv2 YOLOv3 MobileNet-YOLOv3 YOLOv4 YOLOv5 YOLOv7 YOLOX ...
• Detection: NanoDet
• Segmentation: FCN PSPNet UNet YOLACT ...
• Pose Estimation: SimplePose ...

HowTo

• Build for Linux / NVIDIA Jetson / Raspberry Pi3, Pi4 / POWER
• Build for Windows x64 using VS2017
• Build for macOS
• Build for ARM Cortex-A family with cross-compiling
• Build for Hisilicon platform with cross-compiling
• Build for Android
• Build for iOS on macOS with xcode
• Build for WebAssembly
• Build for AllWinner D1
• Build for Loongson 2K1000
• Build for Termux on Android
• Build for QNX

FAQ

Features

• Supports convolutional neural networks, supports multiple input and multi-branch structure, can calculate part of the branch
• No third-party library dependencies, does not rely on BLAS / NNPACK or any other computing framework
• Pure C++ implementation, cross-platform, supports Android, iOS and so on
• ARM NEON assembly level of careful optimization, calculation speed is extremely high
• Sophisticated memory management and data structure design, very low memory footprint
• Supports multi-core parallel computing acceleration, ARM big.LITTLE CPU scheduling optimization
• Supports GPU acceleration via the next-generation low-overhead Vulkan API
• Extensible model design, supports 8bit quantization and half-precision floating point storage, can import caffe/pytorch/mxnet/onnx/darknet/keras/tensorflow(mlir) models
• Support direct memory zero copy reference load network model
• Can be registered with custom layer implementation and extended
• Well, it is strong, not afraid of being stuffed with 卷 QvQ

功能概述

• 支持卷积神经网络，支持多输入和多分支结构，可计算部分分支
• 无任何第三方库依赖，不依赖 BLAS/NNPACK 等计算框架
• 纯 C++ 实现，跨平台，支持 Android / iOS 等
• ARM Neon 汇编级良心优化，计算速度极快
• 精细的内存管理和数据结构设计，内存占用极低
• 支持多核并行计算加速，ARM big.LITTLE CPU 调度优化
• 支持基于全新低消耗的 Vulkan API GPU 加速
• 可扩展的模型设计，支持 8bit 量化 和半精度浮点存储，可导入 caffe/pytorch/mxnet/onnx/darknet/keras/tensorflow(mlir) 模型
• 支持直接内存零拷贝引用加载网络模型
• 可注册自定义层实现并扩展
• 恩，很强就是了，不怕被塞卷 QvQ

supported platform matrix

• ✅ = known work and runs fast with good optimization
• ✔️ = known work, but speed may not be fast enough
• ❔ = shall work, not confirmed
• / = not applied

Project examples

• https://github.com/nihui/ncnn-android-squeezenet
• https://github.com/nihui/ncnn-android-styletransfer
• https://github.com/nihui/ncnn-android-mobilenetssd
• https://github.com/moli232777144/mtcnn_ncnn
• https://github.com/nihui/ncnn-android-yolov5
• https://github.com/xiang-wuu/ncnn-android-yolov7
• https://github.com/nihui/ncnn-android-scrfd 🤩
• https://github.com/shaoshengsong/qt_android_ncnn_lib_encrypt_example

• https://github.com/magicse/ncnn-colorization-siggraph17
  

  

• https://github.com/mizu-bai/ncnn-fortran Call ncnn from Fortran
• https://github.com/k2-fsa/sherpa Use ncnn for real-time speech recognition (i.e., speech-to-text); also support embedded devices and provide mobile Apps (e.g., Android App)

License