K.Kuai Jin

face-anti-spoofing-dataset. Train presentation attack detection and liveness models with this dataset for face anti-spoofing.

skills. Skills for Real Engineers. Straight from my .agents directory.

FunASR. Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

eko. Eko (Eko Keeps Operating) - Build Production-ready Agentic Workflow with Natural Language - eko.fellou.ai

suna. The Company AI Command Center

aptos-core. Aptos is a layer 1 blockchain built to support the widespread use of blockchain through better technology and user experience.

RepViT. RepViT: Revisiting Mobile CNN From ViT Perspective [CVPR 2024] and RepViT-SAM: Towards Real-Time Segmenting Anything

AndroidUSBCamera. 🔥🔥🔥Flexible and useful UVC camera engine on Android platform, supporting multi-road cameras!

ncnn_Android_blazeface. This a demo of BlazeFace infer by ncnn

SAFAS. Code for CVPR 2023 paper "Rethinking Domain Generalization for Face Anti-spoofing: Separability and Alignment"

cvpr2024-face-anti-spoofing-challenge. Accepted by CVPR Workshop 2024

rtrppg. Python implementation of the 3DCNN-based Real-Time rPPG network (RTrPPG)

Face-Anti-Spoofing-Neural-Network. Implementation of "Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervision" in PyTorch.

OC_SORT_CPP. [CVPR 2023] OC-SORT implemented in C++ with Eigen Library, Plus a Android Demo Apk

scrfd-opencv. 使用OpenCV部署SCRFD人脸检测，包含C++和Python两种版本的程序实现，本套程序只依赖opencv库就可以运行， 从而彻底摆脱对任何深度学习框架的依赖。

face-recognition. Real-Time Face Recognition use SCRFD, ArcFace, ByteTrack and Similarity Measure

FaceBagNet. FaceBagNet - Patch-based Methods for Multi-modal Face Anti-spoofing (FAS)

easyLite. easy nosql database for IoT device, clean api and cross platform

zerox. OCR & Document Extraction using vision models

Face_segmentation. Parsing different parts of the face using semantic segmentation.

face_segmentation. Deep face segmentation in extremely hard conditions

SegfaceAndAlignByDlib. 用dlib实现脸部分割和人脸对齐

moondream. tiny vision language model

BitNet. Official inference framework for 1-bit LLMs

ExplainableFaceImageQuality. Pixel-Level Face Image Quality Assessment for Explainable Face Recognition

FaceLib. Face Analysis: Detection, Age Gender Estimation & Recognition

DAT. Repository of Vision Transformer with Deformable Attention (CVPR2022) and DAT++: Spatially Dynamic Vision Transformerwith Deformable Attention

Open-Sora. Open-Sora: Democratizing Efficient Video Production for All

fish-speech. SOTA Open Source TTS

efficientvit. Efficient vision foundation models for high-resolution generation and perception.

ImageNetModel. Official ImageNet Model repository

TransNeXt. [CVPR 2024] Code release for TransNeXt model

Mali_OpenCL_SDK. http://malideveloper.arm.com/resources/sdks/mali-opencl-sdk/

metaformer. MetaFormer Baselines for Vision (TPAMI 2024)

Neighborhood-Attention-Transformer. Neighborhood Attention Transformer, arxiv 2022 / CVPR 2023. Dilated Neighborhood Attention Transformer, arxiv 2022

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

davit. [ECCV 2022]Code for paper "DaViT: Dual Attention Vision Transformer"

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

MegaVIT. The open source implementation of the model from "Scaling Vision Transformers to 22 Billion Parameters"

flash-cosine-sim-attention. Implementation of fused cosine similarity attention in the same style as Flash Attention

ml-aim. This repository provides the code and model checkpoints for AIMv1 and AIMv2 research projects.

LeViT. LeViT a Vision Transformer in ConvNet's Clothing for Faster Inference

lite.ai.toolkit. 🛠A lite C++ AI toolkit: 100+ models with MNN, ORT and TRT, including Det, Seg, Stable-Diffusion, Face-Fusion, etc.🎉

lightqnet-train. Lightweight Face Image Quality Assessment

CvT. This is an official implementation of CvT: Introducing Convolutions to Vision Transformers.

ControlNet. Let us control diffusion models!

few_shot_fas. Python

IADG. (CVPR 2023) Instance-Aware Domain Generalization for Face Anti-Spoofing

DeepFAS. 🔥Deep Learning for Face Anti-Spoofing

ViTransPAD. ViTransPAD: Video Transformer using convolution and self-attention for Face Presentation Attack Detection

ndsplib-hifi5. NatureDSP Library for HiFi5 DSP cores

rknn-toolkit2. C

DECA. DECA: Detailed Expression Capture and Animation (SIGGRAPH 2021)

3DDFA_V2. The official PyTorch implementation of Towards Fast, Accurate and Stable 3D Dense Face Alignment, ECCV 2020.

3DDFA-V3. The official implementation of 3DDFA_V3 in CVPR2024 (Highlight).

Face-Depth-Network. The component of DaGAN (CVPR 2022)

pytorch3d-me. A Pytorch3D extension used in AvatarMe++ and FitMe

AvatarMe. Public repository for the CVPR 2020 paper AvatarMe and the TPAMI 2021 AvatarMe++

MogaNet. [ICLR 2024] MogaNet: Efficient Multi-order Gated Aggregation Network

FasterViT. [ICLR 2024] Official PyTorch implementation of FasterViT: Fast Vision Transformers with Hierarchical Attention

DeepfakeDetection. Deepfake Image Detection with Lightweight Models

DeepSORT. support deepsort and bytetrack MOT(Multi-object tracking) using yolov5 with C++

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

CogVideo. text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

rknn-toolkit2. C

dface. Deep learning face detection and recognition, implemented by pytorch. (pytorch实现的人脸检测和人脸识别)

Blurry-Image-Detector. Robust Python implementation for detecting blurry images using ROI estimation and DCT analysis.

blazeface. face dectection

10kinds-light-face-detector-align-recognition. 10种轻量级人脸检测算法的比拼，其中还包含人脸关键点检测与对齐，人脸特征向量提取和计算距离相似度

tracking.js. A modern approach for Computer Vision on the web

clmtrackr. Javascript library for precise tracking of facial features via Constrained Local Models

Real-World-Masked-Face-Dataset. Real-World Masked Face Dataset，口罩人脸数据集

Mask_face_recognitionZ. Python

RegisterDemo. 一个实现了机器码生成与激活码激活（模拟软件注册）功能的demo

Computer-unique-code-generator. generate the constant & unique code for a computer according to the CPU ID & hard disk id & MAC address & Motherboard ID 根据电脑的 CPU编号、硬盘id编号、MAC地址、主板 id编号 生成一个全球唯一的、持久不变的（除非电脑更换了硬件）、只针对当前电脑的 编码

Dagon_Code. Obtain CPU, motherboard, MAC address, virtual ID (docker, machine-rkt) from Windows, Linux, and generate Dagon_Code.（根据机器指纹生成激活码）

MachineCodeCal. Get the hardware information such cpu id ,cpu sn,mac,motherboard sn etc, then calculate the machine code,this method is useful in Software Registration.

head-pose-estimation. Real-time head pose estimation built with OpenCV and dlib

ParC-Net. [ECCV 2022] Source code of "EdgeFormer: Improving Light-weight ConvNets by Learning from Vision Transformers"

transformer-pytorch. A PyTorch implementation of Transformer in "Attention is All You Need"

transformers. 🤗 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.

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

blazeface_ncnn. 移植pytorch版本blazeface到ncnn框架下

DBFace. DBFace is a real-time, single-stage detector for face detection, with faster speed and higher accuracy

qt-everywhere-src-5.14.2-cross-compile-for-RK3566-RK3568-RK3588. This is the documentation for RK3566 RK3568 RK3588 Tablet and miniPC products

attention-is-all-you-need-pytorch. A PyTorch implementation of the Transformer model in "Attention is All You Need".

transformer. Transformer: PyTorch Implementation of "Attention Is All You Need"

MaskTheFace. Convert face dataset to masked dataset

MobileOne. An Improved One millisecond Mobile Backbone

rknpu_ddk. DDK for Rockchip NPU

webcam-capture. The goal of this project is to allow integrated or USB-connected webcams to be accessed directly from Java. Using provided libraries users are able to read camera images and detect motion. Main project consist of several sub projects - the root one, which contains required classes, build-in webcam driver compatible with Windows, Linux and Mac OS, which can stream images as fast as your camera can serve them (up to 50 FPS). Main project can be used standalone, but user is able to replace build-in driver with different one - such as OpenIMAJ, GStreamer, V4L4j, JMF, LTI-CIVIL, FMJ, etc.

fast-object-pool. A lightweight partitioned object pool, you can use it to pool expensive objects like jdbc connections, thrift clients etc.

ThreadPool. A simple C++11 Thread Pool implementation

cpptime. A portable, pure C++11 timer component.

Fast-MTCNN. a casual work about retraining to optimize mtcnn Pnet and ONet. it can achieve 100+fps on CPU with minSize 60 (1920x1080) on intel i7 6700k

FaceTrack_ncnn_HyperFT. 使用mtcnn和o网络跟踪+光流跟踪进行多目标人脸跟踪，单目标人脸光流跟踪是0.5ms左右

CASIA-SURF_CeFA. Face Anti-spoofing Attack Detection Challenge@CVPR2020

attention-module. Official PyTorch code for "BAM: Bottleneck Attention Module (BMVC2018)" and "CBAM: Convolutional Block Attention Module (ECCV2018)"

SA-Net. Code for our ICASSP 2021 paper: SA-Net: Shuffle Attention for Deep Convolutional Neural Networks

CBAM.PyTorch. Non-official implement of Paper：CBAM: Convolutional Block Attention Module

WaveletSRNet. A pytorch implementation of Paper "Wavelet-srnet: A wavelet-based cnn for multi-scale face super resolution"

run. 润学全球官方指定GITHUB，整理润学宗旨、纲领、理论和各类润之实例；解决为什么润，润去哪里，怎么润三大问题； 并成为新中国人的核心宗教，核心信念。

FaceQuality. An implementation of EQFace: A Simple Explicit Quality Network for Face Recognition (https://arxiv.org/abs/2105.00634, CVPRW 2021)

lightqnet. Deployment of the Lightweight Face Image Quality Assessment

mnn_example. alibaba MNN, mobilenet classifier, centerface detecter, ultraface detecter, pfld landmarker and zqlandmarker, mobilefacenet

Insightface-NPU. C++

CDCN-Face-Anti-Spoofing.pytorch. Apply Central Difference Convolutional Network (CDCN) for face anti spoofing

FaceX-Zoo. A PyTorch Toolbox for Face Recognition

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

BlazeFace-PyTorch. The BlazeFace face detector model implemented in PyTorch

FaceLivenessDetection-SDK. 3D Passive Face Liveness Detection (Anti-Spoofing) & Deepfake detection. A single image is needed to compute liveness score. 99,67% accuracy on our dataset and perfect scores on multiple public datasets (NUAA, CASIA FASD, MSU...).

FAS-SGTD. Deep Spatial Gradient and Temporal Depth Learning for Face Anti-spoofing

CelebA-Spoof. [ECCV2020] A Large-Scale Face Anti-Spoofing Dataset

PaddleDetection. Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.

openwifi. open-source IEEE 802.11 WiFi baseband FPGA (chip) design: driver, software

taskt. taskt (pronounced 'tasked' and formely sharpRPA) is free and open-source robotic process automation (rpa) built in C# powered by the .NET Framework

Airtest. UI Automation Framework for Games and Apps

FlaUI. UI automation library for .Net

Rehosted-Workflow-Designer. .NET Workflow Designer Rehosting & Custom Activity Library Example

rpaframework. Collection of open-source libraries and tools for Robotic Process Automation (RPA), designed to be used with both Robot Framework and Python

opencore. Scalable, secure workflow solution

RPAStudio. RPA机器人流程自动化

micronet. micronet, a model compression and deploy lib. compression: 1、quantization: quantization-aware-training(QAT), High-Bit(>2b)(DoReFa/Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference)、Low-Bit(≤2b)/Ternary and Binary(TWN/BNN/XNOR-Net); post-training-quantization(PTQ), 8-bit(tensorrt); 2、 pruning: normal、regular and group convolutional channel pruning; 3、 group convolution structure; 4、batch-normalization fuse for quantization. deploy: tensorrt, fp32/fp16/int8(ptq-calibration)、op-adapt(upsample)、dynamic_shape

awesome-zero-shot-learning. A curated list of papers, code and resources pertaining to zero shot learning

Knowledge-Distillation-Zoo. Pytorch implementation of various Knowledge Distillation (KD) methods.

tarball-lite. AFAD-LITE Tarball

tarball. AFAD Tarballs

native-utils. A simple library class which helps with loading dynamic JNI libraries stored in the JAR archive

SSR-Net. [IJCAI18] SSR-Net: A Compact Soft Stagewise Regression Network for Age Estimation

MaskInsightface. 基于人脸关键区域提取的人脸识别（LFW:99.82%+ CFP_FP:98.50%+ AgeDB30:98.25%+）

facerecognize-for-mobile-phone. 适用于移动端的人脸识别模型，计算量与mobilefacenet相同，但megaface上提升了2%+

rknn-toolkit. Python

TNN. TNN: developed by Tencent Youtu Lab and Guangying Lab, a uniform deep learning inference framework for mobile、desktop and server. TNN is distinguished by several outstanding features, including its cross-platform capability, high performance, model compression and code pruning. Based on ncnn and Rapidnet, TNN further strengthens the support and performance optimization for mobile devices, and also draws on the advantages of good extensibility and high performance from existed open source efforts. TNN has been deployed in multiple Apps from Tencent, such as Mobile QQ, Weishi, Pitu, etc. Contributions are welcome to work in collaborative with us and make TNN a better framework.

webrtc. Pure Go implementation of the WebRTC API

SRGAN. A PyTorch implementation of SRGAN based on CVPR 2017 paper "Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network"

OUCML. Python

LFFD-A-Light-and-Fast-Face-Detector-for-Edge-Devices. A light and fast one class detection framework for edge devices. We provide face detector, head detector, pedestrian detector, vehicle detector......

CPlusPlusThings. C++那些事

drawio-desktop. Official electron build of draw.io

frvt. Repository for the Face Recognition Vendor Test (FRVT)

AES-ARM-NEON. Efficient implementation of maksed AES on ARM NEON

pyAudioAnalysis. Python Audio Analysis Library: Feature Extraction, Classification, Segmentation and Applications

pytorch-gradual-warmup-lr. Gradually-Warmup Learning Rate Scheduler for PyTorch

loris. Loris is a library for sound analysis, synthesis, and morphing, developed by Kelly Fitz and Lippold Haken at the CERL Sound Group. Loris includes a C++ class library, Python module, C-linkable interface, command line utilities, and documentation.

LFFD-with-ncnn. A C++ API of the LFFD with ncnn

PFLD_ncnn_test. for calculating euler angle test in ncnn

oneTBB. oneAPI Threading Building Blocks (oneTBB)

neon-guide. Makes ARM NEON documentation accessible (with examples)

Real-Time-Person-Removal. Removing people from complex backgrounds in real time using TensorFlow.js in the web browser

jittor. Jittor is a high-performance deep learning framework based on JIT compiling and meta-operators.

RxAndroid. RxJava bindings for Android

AndroidDevWithCpp. Android Develop With C++

chineseocr_lite. 超轻量级中文ocr，支持竖排文字识别, 支持ncnn、mnn、tnn推理 ( dbnet(1.8M) + crnn(2.5M) + anglenet(378KB)) 总模型仅4.7M

test3.

connect4. connect4, Cricket, alphagozero, Thompson bound MCTS, KL-UCB bound MCTS, UCB bound MCTS

skiprnn_pytorch. A pytorch implementation of the paper: "Skip RNN: Learning to Skip State Updates in Recurrent Neural Networks"

deep-high-resolution-net.pytorch. The project is an official implementation of our CVPR2019 paper "Deep High-Resolution Representation Learning for Human Pose Estimation"

Riptide. Simple Training and Deployment of Fast End-to-End Binary Networks

PFLD-pytorch. PFLD pytorch Implementation

pytorch-lightning. Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

opencv. Open Source Computer Vision Library

handtracking. Building a Real-time Hand-Detector using Neural Networks (SSD) on Tensorflow

project_kojak. Training a Neural Network to Detect Gestures and Control Smart Home Devices with OpenCV in Python

pfld-pytorch. Python

PFLD. Implementation of PFLD A Practical Facial Landmark Detector , reference to https://arxiv.org/pdf/1902.10859.pdf

lffd-pytorch. A light and fast one class detection framework for edge devices. We provide face detector, head detector, pedestrian detector, vehicle detector......

FaceMaskDetection. 开源人脸口罩检测模型和数据 Detect faces and determine whether people are wearing mask.

ncnn-ssd-face-detection. ncnn face detection using depthwise convolution

ssh_detector. c++ implementation for ssh detector for object detect. something likes ssd

crnn-audio-classification. UrbanSound classification using Convolutional Recurrent Networks in PyTorch

crnn-pytorch. Pytorch implementation of OCR system using CRNN + CTCLoss

fate. Fate is a modern science chinese name create tool.

horovod. Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

pytorch-cifar100. Practice on cifar100(ResNet, DenseNet, VGG, GoogleNet, InceptionV3, InceptionV4, Inception-ResNetv2, Xception, Resnet In Resnet, ResNext,ShuffleNet, ShuffleNetv2, MobileNet, MobileNetv2, SqueezeNet, NasNet, Residual Attention Network, SENet, WideResNet)

senet.pytorch. PyTorch implementation of SENet

ocr_densenet. 第一届西安交通大学人工智能实践大赛（2018AI实践大赛--图片文字识别）第一名；仅采用densenet识别图中文字

codelite. A multi purpose IDE specialized in C/C++/Rust/Python/PHP and Node.js. Written in C++

librosa. Python library for audio and music analysis

onnxruntime. ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

armnn. Arm NN ML Software.

honk2. PyTorch implementations of neural network models for keyword spotting

pytorch-kaldi. pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit.

ASRFrame. An Automatic Speech Recognition Frame ，一个中文语音识别的完整框架， 提供了多个模型

EfficientNet-PyTorch. A PyTorch implementation of EfficientNet

patch-depth-based-anti-spoofing. Python

AluminiumDefectRecognition. 天池比赛-铝型材表面瑕疵识别的实现；Pytorch；finetune renet;

rPPG. Python

Face-Anti-spoofing. Face Anti-spoofing demo to detect print, video replay, and mask attack by using LBP features and rPPG features.

rppg-pos. Remote photoplethysmography measures heart rate of a person without any contact, from their video. This is an implementation of rPPG called as Plane Orthogonal-to-Skin (POS) as described in the IEEE paper - "Algorithmic Principles of Remote PPG," W. Wang, A. C. den Brinker, S. Stuijk and G. de Haan.

boxx. Tool-box for efficient build and debug in Python. Especially for Scientific Computing and Computer Vision.

heartbeat. Desktop implementation of Remote Photoplethysmography – Measuring heart rate using facial video.

2DASL. The code (pytorch for testing & matlab for 3D plot and evaluation) for our project: Joint 3D Face Reconstruction and Dense Face Alignment from A Single Image with 2D-Assisted Self-Supervised Learning （2DASL）

3DDFA. The PyTorch improved version of TPAMI 2017 paper: Face Alignment in Full Pose Range: A 3D Total Solution.

pytorch-cnn-finetune. Fine-tune pretrained Convolutional Neural Networks with PyTorch

roam. Implementation of the state-of-the-art video-segmentation / rotoscoping tool described in the CVPR 2017 paper ROAM: a Rich Object Appearance Model with Application to Rotoscoping

buildbot. Python-based continuous integration testing framework; your pull requests are more than welcome!

ML-KWS-for-FPGA. https://github.com/ARM-software/ML-KWS-for-MCU

SimpleCV. The Open Source Framework for Machine Vision

ColorSpace. C++ library for working with different color spaces, converting between color spaces and comparing colors

ML-KWS-for-MCU. Keyword spotting on Arm Cortex-M Microcontrollers