Skip to content

EdgeDevice.md

Latest commit

 

History

History
113 lines (75 loc) · 4.88 KB

EdgeDevice.md

File metadata and controls

113 lines (75 loc) · 4.88 KB

OpenEdge：开放的边缘计算平台，将云计算能力拓展至用户现场，提供临时离线、低延时的计算服务，包括消息路由、函数计算、AI推断等 https://github.com/baidu/openedge

在Arm Cortex CPUs/Arm Mali GPUs上跑主流深度学习开发框架的开发包 https://github.com/ARM-software/armnn

AI Benchmark: Running Deep Neural Networks on Android Smartphones http://ai-benchmark.com/

利用Mobilenetssd目标检测框架,ncnn前向推理,android项目 https://github.com/chehongshu/ncnnforandroid_objectiondetection_Mobilenetssd

ARM GPU等边缘计算设备上的深度学习视频监控、人脸识等 https://github.com/SharpAI/DeepCamera

MNN Implementation of the paper of LFFD: A Light and Fast Face Detector for Edge Devices

https://github.com/SyGoing/LFFD-MNN

A C++ API of the LFFD with ncnn

https://github.com/SyGoing/LFFD-with-ncnn

【TinyML：边缘微机器学习工程】 https://www.informationweek.com/big-data/ai-machine-learning/engineering-tiny-machine-learning-for-the-edge/a/d-id/1336972

【嵌入(Embedding-Based)机器学习模型部署通用模式】

https://medium.com/m/global-identity?redirectUrl=https%3A%2F%2Fblog.ml6.eu%2Fa-general-pattern-for-deploying-embedding-based-machine-learning-models-bf12e8979070

【TensorRT 加速实现的实时姿态估计】

https://github.com/NVIDIA-AI-IOT/trt_pose

5MB的神经网络也高效，Facebook新压缩算法造福嵌入式设备 https://mp.weixin.qq.com/s/m9I5TM9uJcgZvMusO667OA

'超迷你智能摄像头A-Eye - 基于K210的超迷你开发板，软硬件完全开源' https://github.com/david-pzh/A-Eye

【嵌入式AI(边缘AI)相关资源列表】 https://github.com/crespum/edge-ai

【Bolt：面向移动设备的高性能、异构灵活的深度学习框架】 https://github.com/huawei-noah/bolt

物联网相关项目/资源大列表 https://github.com/HQarroum/awesome-iot

AI Transformer：将(Keras)深度学习模型转换成可用于嵌入式环境的可读无依赖C程序 https://www.aitransformer.com/

边缘机器学习相关资源大列表 https://github.com/Bisonai/awesome-edge-machine-learning

AWTK = Toolkit AnyWhere(为嵌入式、手机和PC打造的通用GUI系统) https://github.com/zlgopen/awtk

AoE (AI on Edge，终端智能，边缘计算) ：滴滴开源的 终端侧 AI 集成运行时环境 ( IRE )，以 “稳定性、易用性、安全性” 为设计原则 https://didi.github.io/AoE/ https://github.com/didi/AoE

嵌入式/物联网安全相关资源大列表 https://github.com/fkie-cad/awesome-embedded-and-iot-security

面向ARM设备的非官方PyTorch/torchvision编译版 https://github.com/nmilosev/pytorch-arm-builds

OpenBot: 用安卓智能手机DIY低成本开源机器人(小车)

https://github.com/intel-isl/OpenBot

pibooth：用树莓派+OpenCV自制“智能”自拍亭 https://github.com/pibooth/pibooth

如何在ARM开发板上从源码编译安装OpenCV和OpenCV contrib https://mp.weixin.qq.com/s?__biz=MjM5NTE3NjY5MA==&mid=2247485240&idx=1&sn=6025fa13cf2b88a356e55c3fdc2b75f1&chksm=a6fdcea5918a47b3d132281e9dfeb31e0f5a8f16f87ab3a5c4defebe4eea28689d13963d9aec&token=1226895782&lang=zh_CN#rd

NNtrainer：嵌入式设备上训练神经网络模型的软件框架 https://github.com/nnstreamer/nntrainer

Visual Imitation Made Easy https://github.com/sarahisyoung/Visual-Imitation-Made-Easy

Mcunet: Tiny deep learning on iot devices

https://github.com/mit-han-lab/tinyml

AI-on-the-edge-device：边缘设备AI应用实例——远程读表

https://github.com/jomjol/AI-on-the-edge-device

deepC：独立于供应商的深度学习库、编译器和推理框架，专为小型设备设计，包括μ控制器、物联网和边缘设备等 https://github.com/ai-techsystems/deepC

用稀疏推理加速移动端和网页上的神经网络 https://ai.googleblog.com/2021/03/accelerating-neural-networks-on-mobile.html

Forward：腾讯开源的 GPU 高性能推理加速框架，可直接加载主流框架模型（Tensorflow / PyTorch / Keras）转换成 TensorRT 推理加速引擎 github.com/Tencent/Forward

tinyMLPerf Deep Learning Benchmarks for Embedded Devices：嵌入式设备tinyML深度学习基准 github.com/mlcommons/tiny

'TensorFlow Lite, Coral Edge TPU samples (Python/C++, Raspberry Pi/Windows/Linux).' github.com/NobuoTsukamoto/edge_tpu

Edge ML Library：高性能端侧机器学习计算库 github.com/netease-youdao/EMLL

PyTorch-LIT：终端设备大模型快速推理工具包 github.com/AminRezaei0x443/PyTorch-LIT

BED: 边缘设备实时目标检测系统

github.com/datamllab/BED_main

【ML Examples：Arm Software的机器学习教程和实例】’ML Examples - Arm Machine Learning tutorials and examples' by Arm Software GitHub: github.com/ARM-software/ML-examples

'TPG: TensorRT Plugin Generator - A simple tool that can generate TensorRT plugin code quickly.' by NVIDIA AI IOT GitHub: github.com/NVIDIA-AI-IOT/tensorrt_plugin_generator