landmarks-detection

This repository contains code for collecting pose data of various yoga poses using the MediaPipe Pose model. The collected data includes the angles of different body joints in each yoga pose.

Fast face detection, pupil detection and landmarks for Android without dependencies.

This project is focused on Action Recognition for Sign Language, which aims to bridge the communication gap between sign language users and those who may not understand sign language.

Final project of CSCI 1430 with a self-implemented face landmark detection model. Implemented a real-time face swap software which can change user’s face with different masks.

[yolov5 v7.0][WIDER FACE][ECCV 2022]YOLO5Face: Why Reinventing a Face Detector

ASL gesture recognition system using MediaPipe, computer vision and machine learning to convert signs into text and speech.

Master thesis at Wroclaw University of Science and Technology.

A Quality Control Model for Wrist X-ray Images

An AI-powered real-time hand gesture recognition system built with Python, MediaPipe, OpenCV, and TensorFlow Lite. The project detects 21 hand landmarks through a webcam, recognizes static hand signs and dynamic finger movements, and displays live gesture predictions with landmark tracking and FPS and supports custom gesture data collection.

2D Facial Landmarks Detection using a convolutional neural network on AFLW2000.

This project is about drawing on air with hand movements and recognize gestures using opencv and mediapipe.