FastTracker: Real-Time and Accurate Visual Tracking

An Embeddable General Purpose Macro Processor

The official code for our ECCV22 oral paper: tracking objects as pixel-wise distributions.

workbench for learning and practicing on-device AI technology in real scenario with online-TV on Android phone, powered by ggml(llama.cpp,whisper.cpp...) and FFmpeg and opencv-mobile

Object Tracking Library

🐼 One Millisecond Deformable Shape Tracking Library (DEST)

A Keras + TensorFlow implementation of the CVPR 2016 paper "Eye Tracking for Everyone"

Compressed-Domain Video Object Tracking using Markov Random Fields with Graph Cuts Optimization

Implemented and improved the iTracker model proposed in the paper "Eye Tracking for Everyone"

Lightweight Visual Tracking: A multipurpose visual odometry system.

The project is an C++ implementation of Face Alignment at 3000fps via Regressing Local Binary Features

MobileGaze: A compact vision framework for mobile gaze tracking

C++ implementation of RemoteEye

106-point face landmarks

Real time eye tracking for embedded and mobile devices.

Python tools for 3D face: 3DMM, Mesh processing(transform, camera, light, render), 3D face representations.

Supervision-by-Registration: An Unsupervised Approach to Improve the Precision of Facial Landmark Detectors

code of BMVC2019 paper 'Photoplethysmograph Signal Measurement from Facial Videos Using Spatio-Temporal Networks'

A deep learning based gaze estimation framework implemented with PyTorch

Facial Landmark Detection based on PyTorch

106点人脸关键点检测的PFLD算法实现

Computer Vision library for human-computer interaction. It implements Head Pose and Gaze Direction Estimation Using Convolutional Neural Networks, Skin Detection through Backprojection, Motion Dete…

Gaze Estimation via Deep Neural Networks

MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire:

my unofficial implementation of PFLD paper "A Practical Facial Landmarks Detector" for a real time landmarks detection & head pose estimation

PFLD实现Gaze Estimation

LWCC: A LightWeight Crowd Counting library for Python that includes several pretrained state-of-the-art models.

DM-CountのONNX変換/推論のサンプル