Home

Awesome


日志

  1. 针对mtcnn的col-major模型添加了caffe2ncnn_mtcnn.cpp, 生成row-major模型
  2. 添加void resize_image(ncnn::Mat& srcImage, ncnn::Mat& dstImage),无arm优化
  3. 前面都是用BGR图像做的,导致很多漏检,改过来了。
  4. 3519编译:在CMakelist.txt的17行下面手动添加如下代码
set(CMAKE_C_COMPILER   arm-hisiv500-linux-gcc) 
set(CMAKE_CXX_COMPILER arm-hisiv500-linux-g++)
add_definitions(-D__ARM_NEON)
add_definitions("-O3 -mfloat-abi=softfp -mfpu=neon-vfpv4 -ffunction-sections") 
  1. 在非手机的arm环境下,需要修改CMakeLists.txt才能把arm编译进去,如CMakeLists_arm.txt
  2. 如果使用opencv2,需要将examples/CmakeLists.txt里的imgcodecs去掉,否则可能编译不过去
  3. 在tx1下make时可能会出现"can't find -lopencv_dep_cudart",重新cmake,"cmake -D CUDA_USE_STATIC_CUDA_RUNTIME=OFF .."
  4. 新版本ncnn,使用mtcnn/mtcnn_new.cpp

image

ncnn

License Build Status

ncnn is a high-performance neural network inference computing framework optimized for mobile platforms. ncnn is deeply considerate about deployment and uses on mobile phones from the beginning of design. ncnn does not have third party dependencies, it is cross-platform, and runs faster than all known open source frameworks on mobile phone cpu. Developers can easily deploy deep learning algorithm models to the mobile platform by using efficient ncnn implementation, create intelligent APPs, and bring the artificial intelligence to your fingertips. ncnn is currently being used in many Tencent applications, such as QQ, Qzone, WeChat, Pitu and so on.

ncnn 是一个为手机端极致优化的高性能神经网络前向计算框架。ncnn 从设计之初深刻考虑手机端的部署和使用。无第三方依赖,跨平台,手机端 cpu 的速度快于目前所有已知的开源框架。基于 ncnn,开发者能够将深度学习算法轻松移植到手机端高效执行,开发出人工智能 APP,将 AI 带到你的指尖。ncnn 目前已在腾讯多款应用中使用,如 QQ,Qzone,微信,天天P图等。


HowTo

how-to-build-ncnn-library

how-to-use-ncnn-with-alexnet

ncnn 组件使用指北 alexnet


Features

功能概述


Example project

Android and iOS https://github.com/dangbo/ncnn-mobile (Thanks to dangbo)

技术交流QQ群:637093648 答案:卷卷卷卷卷


License

BSD 3 Clause