nihui

ncnn is a high-performance neural network inference framework optimized for the mobile platform

waifu2x converter ncnn version, runs fast on intel / amd / nvidia / apple-silicon GPU with vulkan

The minimal opencv for Android, iOS, ARM Linux, Windows, Linux, MacOS, HarmonyOS, WebAssembly, watchOS, tvOS, visionOS

RIFE, Real-Time Intermediate Flow Estimation for Video Frame Interpolation implemented with ncnn library

Detect CPU features with single-file

real-cugan converter ncnn version, runs fast on intel / amd / nvidia / apple-silicon GPU with vulkan

RealSR super resolution implemented with ncnn library

The YOLOv5 object detection android example

DAIN, Depth-Aware Video Frame Interpolation implemented with ncnn library

ncnn android nanodet lightweight object detection

ncnn implementation of Z-Image image generater

SRMD super resolution implemented with ncnn library

No description.

ncnn android scrfd face detection

ncnn android yolov8 realtime detection, segmentation, pose estimation, classification and obb

The squeezenet image classification android example

ncnn android paddle ocr v5

A tool which profiles Vulkan devices to find their peak capacities

ncnn benchmark on various single board computers

Deploy YOLOv5 in your web browser with ncnn and webassembly

CAIN, Channel Attention Is All You Need for Video Frame Interpolation implemented with ncnn library

ncnn android yolo11 realtime detection, segmentation, pose estimation, classification and obb

Deploy nanodet, the super fast and lightweight object detection, in your web browser with ncnn and webassembly

The style transfer android example

IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation implemented with ncnn library

The mobilenetssd object detection android example

ncnn android benchmark app

No description.

ncnn android piper the fast and local neural text-to-speech engine

No description.

No description.

No description.

mesa turnip driver for android

Deploy SCRFD, an efficient high accuracy face detection approach, in your web browser with ncnn and webassembly

Portrait segmentation in your web browser with ncnn and webassembly

ncnn android vkpeak

ncnn android robust video matting

No description.

prebuild package for cross compiling riscv

No description.

No description.

No description.

Real-time AI video segmentation of USB camera and streaming over HTTP

QRCode(from WeChat) implement in ncnn⚡二维码检测&解码⚡ncnn⚡

No description.

No description.

A repo for llm on ncnn

🎮 A step-by-step guide to implementing SSAO, depth of field, lighting, normal mapping, and more for your 3D game.

⚡Super fast and lightweight anchor-free object detection model. 🔥Only 1.8mb and run 97FPS on cellphone🔥

Visualizer for neural network, deep learning and machine learning models

The General Purpose Vulkan Compute Framework. Blazing fast, lightweight, mobile-enabled, and optimized for advanced GPU data processing usecases.

Real-time speech recognition using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Raspberry Pi, VisionFive2, etc.

add custom op in both tensorflow and tensorflow-lite

1000 images, one per image-net class. For easy visualization/exploration of classes.

超轻量级中文ocr，支持竖排文字识别, 支持ncnn推理 , dbnet(1.7M) + crnn(6.3M) + anglenet(1.5M) 总模型仅10M

No description.

Lightweight Mat and imread()/imwrite()/imshow().

No description.

OpenCL for Nets - A Deep Learning Framework based on OpenCL, written by C++. Supports popular MLP, RNN(LSTM), CNN. Friendly Debugger. Transparent Data. 基于OpenCL的深度学习计算框架，C++开发，支持多层感知器，长短时记忆模型，卷积神经网络。调试方便，数据透明。

No description.

A fast desktop environment based on KDE technology

No description.

Simplify your onnx model

Khronos-reference front end for GLSL/ESSL, partial front end for HLSL, and a SPIR-V generator.

This project include face-deteciton face-landmarks face-align and face-recgnition; It is a relatively complete project！

NEW - YOLOv8 🚀 in PyTorch > ONNX > OpenVINO > CoreML > TFLite

No description.

Mobile AI Compute Engine

No description.

一个移动端跨平台的gpu+cpu并行计算的cnn框架

A Python FFI of nihui/rife-ncnn-vulkan achieved with SWIG

A Toolkit to Help Optimize Large Onnx Model

Convert between RGB and YUV

TNN：移动端高性能、轻量级推断框架，同时拥有跨平台、高性能、模型压缩、代码裁剪等众多突出优势。TNN框架在原有Rapidnet、ncnn框架的基础上进一步加强了移动端设备的支持以及性能优化，同时也借鉴了业界主流开源框架高性能和良好拓展性的优点。目前TNN已经在手Q、微视、P图等应用中落地，欢迎大家参与协同共建，促进TNN推断框架进一步完善。

EAGLEEYE图像应用开发框架

nniefacelib是一个在海思35xx系列芯片上运行的人脸算法库

Vulkan On Metal

No description.

some tools to transform darknet/tensorflow models to tencent ncnn.

xf86 Driver for SIS Chipset

A build project for ONNX Runtime

No description.

vits Android部署

Tencent NCNN with added CUDA support

Source code for Dain-App

Tengine is a lite, high performance, modular inference engine for embedded device

No description.

Bolt is a deep learning framework with high performance and heterogeneous flexibility.

dabnn is an accelerated binary neural networks inference framework for mobile platform

MNN is a lightweight deep neural network inference engine.

This a bridge for converting torch,and other AI training framework to C++ speed up infer library,like NCNN and ect

Tensors and Dynamic neural networks in Python with strong GPU acceleration

基于opengles的神经网络前向传播框架

Darknet2ncnn converts the darknet model to the ncnn model

Lightweight ncnn C API bindings for MicroPython

No description.

Real-World Super-Resolution via Kernel Estimation and Noise Injection

No description.

next generation of ofetion library

This is a Experimental version of OpenCL by AMD Research, we now recommend you to use The official BVLC Caffe OpenCL branch is over at Caffe branch now at https://github.com/BVLC/caffe/tree/opencl

This convert tools is base on TensorRT 2.0 Int8 calibration tools,which use the KL algorithm to find the suitable threshold to quantize the activions from Float32 to Int8(-128 - 127).

Implement streaming NCNN C++ version of FireRedVAD

:zap: A newly designed ultra lightweight anchor free target detection algorithm， weight only 250K parameters， reduces the time consumption by 10% compared with yolo-fastest, and the post-processing is simpler

Face Analysis Project on MXNet and PyTorch

ncnnRay++ is a CMake based integration of raylib and the very popular Tencent ncnn Deep Learning library. ncnn is written in C++ designed (but not only) for edge computing devices. The project depends on the Vulkan SDK (Vulakn is Khronos' API for Graphics and Compute on GPUs).

No description.

ONNX Runtime: cross-platform, high performance scoring engine for ML models

No description.

pytorch 2 caffe

No description.

Awesome Vulkan ecosystem

FeatherCNN is a high performance inference engine for convolutional neural networks.

Benchmarking Neural Network Inference on Mobile Devices

An Embedded Computer Vision & Machine Learning Library (CPU Optimized & IOT Capable)

No description.

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more

ONNX Script enables developers to naturally author ONNX functions and models using a subset of Python.

Porting caffe opencl to android platform

No description.

Tengine is a lite, high performance, modular inference engine for embedded device

Minimal runtime core of Caffe, Forward only, GPU support and Memory efficiency.

Protocol Buffers - Google's data interchange format

No description.

一个测试版

Unofficial MIPS Loongson Intrinsics Guide

eSpeak NG is an open source speech synthesizer that supports more than hundred languages and accents.

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.

Submission repo for s2309. ref: rvspoc.org

OS (Operating System) logos list in different sizes. This package contains +80 OS logos and it helps statistic or log listing and graph visualizing.

No description.

No description.

No description.

TinyMaix is a tiny inference library for microcontrollers (TinyML).

No description.

A primitive library for neural network

Datasets, Transforms and Models specific to Computer Vision

No description.

PyTorch Neural Network eXchange

🎡 Build Python wheels for all the platforms on CI with minimal configuration.

yolort / yet another yolov5, with its runtime stack for libtorch, onnx, tvm, ncnn and specialized accelerators.

Open deep learning compiler stack for Kendryte K210 AI accelerator

No description.

A read-only Git copy of the OpenBenchmarking.org test profiles.

Tengine Convert Tool supports converting multi framworks' models into tmfile that suitable for Tengine-Lite AI framework.

B站直播彈幕工具

Vulkan Loader

Source code for AAAI 2020 paper "Channel Attention Is All You Need for Video Frame Interpolation"

A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.

No description.

The RISC-V software tools list, as seen on riscv.org

Language-specific restylers

An Open Source Machine Learning Framework for Everyone

No description.

No description.

No description.

No description.

No description.

No description.

No description.

Quantized Neural Network PACKage - mobile-optimized implementation of quantized neural network operators

R-C3D pytorch implementation

nihui's patch collection

No description.

Caffe: a fast open framework for deep learning.

a sampling from ImageNet: 5 examples for each of 200 classes

No description.

Vulkan On D3D12

No description.

More advanced badges for projects using Travis or Sauce Labs

NyaaWiki, built with Docsify.

common

No description.

No description.

No description.

No description.

No description.

No description.

No description.

Reverse engineering the V831 npu

Unofficial LoongArch Intrinsics Guide

Termux for HarmonyOS Computer. Working in Progress.

Github action to download and install a specific version of the Windows 10 SDK 📦

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

MoltenVK is a Vulkan Portability implementation. It layers a subset of the high-performance, industry-standard Vulkan graphics and compute API over Apple's Metal graphics framework, enabling Vulkan applications to run on macOS, iOS and tvOS.

No description.

SDK for IDO-SBC2D06, base on SSD20X V30 SDK.

ncnn sample runs end to end (zero copy) on android (gets camera, processes with ncnn, renders in window)

No description.

MC3172-CMake Project

Android paddleocr demo infer by ncnn

No description.

No description.

试图整理汇总 MtF 的相关资料，为大家提供更好的帮助~

No description.

Repository for the t2linux.org wiki

YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported.

the C++ version of SeqSeq with ncnn

Google Brain AutoML

ABBS/ACBS tree for AOSC OS package metadata, build configuration, scripts, and patches

Command-line program to download videos from YouTube.com and other video sites