package google mediapipe hand and holistic tracking into a dynamic link library

A MediaPipe Solver library, converting Moetion (inspired by Kalidokit) for use in Unity

Single camera full body motion capture. Drive 3D humanoid models right from your browser.

This project integrates MediaPipe Solutions with Node.js and Express for real-time computer vision tasks. It showcases examples of image segmentation, hand and face detection, and pose detection, with a combined example for all three types of landmark detection.

YOLOv5-OpenPose designed for CPU

A Flask web application game that detects real time Dynamic Sign Language using LSTM Keypoint Detection Model.

Emotion based music recommender system

A COMPUTER VISION PROJECT USING GOOGLE MEDIAPIPE

Gear is a WIP motion-based minigame.

The proposed system uses a Long Short-Term Memory (LSTM) model with Mediapipe holistics which detects landmarks on hand for detection of hand gestures which helps in recognizing and deciphering British sign language (BSL) in Real Time. In BSL both hands are used for gesturing.

A web application powered by machine learning models that predict the yoga asana performed by users. Designed to enhance the daily practice of yoga enthusiasts, this application offers real-time feedback on posture, enabling practitioners to refine their technique, elevate their practice, and mitigate the risk of injuries.

Bangla sign recognition system is developed based on a deep learning approach named LSTM. As a feature extractor, Mediapipe Holistic is also used in this system. This system is able to work on static and dynamic both types of gestures and recognize simple or complex signs.

A keypoint detection model consisting of an action detection model to decode certain actions. This is built using Tensorflow, keras to build a deep neural network introducing LSTM layers to handle the sequence of keypoints using MediaPipe Holistic Keypoints

A fullstack gesture recognition project that utilises Mediapipe and LSTM Neural Networks to recognise gestures in real time captured by device camera using OpenCV and pay way to enable IoT device control using those gestures in the future iterations.

Public Repository for the Project done in the MMS Cours of the fall semester 2023 at the University of Bern

Ever just wanted an excuse to leave that awkward date/party? With Wing, we can give you the perfect excuse to get out by getting a phantom call from your mom whenever you scratch your head

Bachelor's degree project. LSTM model that learns Sign language and translates it to subtitles on the fly.

Holistic detection (tracking face, eye and hand) with an excel output

Real-time calibrated multi-camera 3D coordinate capture system featuring intrinsic/extrinsic camera calibration, multithreaded parallel processing, and robust markerless multi-person detection and tracking using YOLOv11, DeepSort, and MediaPipe Holistic for precise 3D landmark extraction across synchronized camera views.