The deaf-mute community have undeniable communication problems in their daily life. Recent developments in artificial intelligence tear down this communication barrier. The main purpose of this paper is to demonstrate a methodology that simplified Sign Language Recognition using MediaPipe’s open-source framework and machine learning algorithm. The predictive model is lightweight and adaptable to smart devices. Multiple sign language datasets such as American, Indian, Italian and Turkey are used for training purpose to analyze the capability of the framework. With an average accuracy of 99%, the proposed model is efficient, precise and robust. Real-time accurate detection using Support Vector Machine (SVM) algorithm without any wearable sensors makes use of this technology more comfortable and easy.

Gesture recognition using Mediapipe Framework and KNN algorithm.

This project extracts feature points based on the Mediapipe open-source model, uses LSTM, CNN, and Vision of Transformer network models for training to obtain a deep learning model, and finally deploys and applies it.

Training a controlnet using mediapipe landmarks with huggingface models and datasets.

Action Recognition Framework: A Python-based framework for generating action recognition data and training models using OpenCV, numpy, PyTorch, and MediaPipe. It supports generating training data, training models with LSTM and Transformer architectures, and real-time action recognition inference

Real-time ASL hand sign recognition using MediaPipe + Machine Learning + Motion Tracking (J/Z). Includes dataset collection, model training, live demo, and smart word suggestions.

This project aims to bridge communication gaps for non-verbal individuals by translating sign language gestures into real-time English captions using deep learning. The application leverages MediaPipe for gesture recognition, TensorFlow for model training, and Streamlit for a user-friendly web interface. Users can record gestures via webcam and see

SignifyAI is a deep learning–based project that uses computer vision to recognize hand gestures and map them to emojis. It includes data collection, preprocessing, model training, and real-time inference using OpenCV and Mediapipe.

This repository presents a real‑time emotion recognition pipeline leveraging MediaPipe for landmark extraction and a hybrid CNN‑BiLSTM‑Attention model. It features CSV data capture, preprocessing with augmentation, class‑weighted training, and live webcam inference with optional landmark visualization for robust affective computing

Real-Time ASL Static Gesture Recognition using MediaPipe & Lightweight ML Models (SVM/MLP) A high-performance ASL translator that runs fully on CPU using MediaPipe hand landmarks and optimized ML models. Includes dataset collection, training pipeline, real-time inference, visualization, and performance metrics.

A real-time sign language translator using Python, MediaPipe, and TensorFlow. Translates gestures to text/speech via webcam, with customizable model training for accuracy and an intuitive interface.

The project aims to develop a real-time gesture recognition system using computer vision techniques and machine learning. The system utilizes the Mediapipe library for pose estimation and TensorFlow for training a deep learning model to recognize gestures.

This project builds a real-time sign language recognition system using deep learning (1DCNNs & Transformers) and MediaPipe hand landmarks. It allows users to fingerspell letters/numbers or express signs for real-time translation. Pre-trained models and scripts for training/inference are included.

DL project on the topic 'Sports Training Assistant App' using Mediapipe, OpenCV, and LLM models (Gemini AI).

Yoga pose classifier with MediaPipe keypoint detection, custom PyTorch model training, and real-time video processing via Streamlit.

Train your own custom hand gestures using Mediapipe & OpenCV. Real-time recognition with a clean Tkinter GUI and ML model training built-in.

A real-time Python application for American Sign Language gesture recognition, using TensorFlow for model training, MediaPipe for hand detection, and OpenCV for image manipulation.

A complete face recognition security system using LBPH, MediaPipe, Flask, and Docker. Includes dataset capture, model training, threshold testing, real-time prediction, and a browser-based authentication UI.

Dynamic Hand Gesture Recognition using MediaPipe & LSTM This project implements real-time dynamic hand gesture recognition using MediaPipe for hand keypoint detection and an LSTM-based deep learning model for gesture classification. It supports training on custom datasets and provides live webcam-based gesture recognition.

Real-time ASL hand-sign recognition with MediaPipe landmarks + Random Forest, and on-frame multilingual translation (via GoogleTrans). Includes data capture, dataset building, model training, and live inference with overlays.

A Python-based ISL/ASL Hand Gesture Recognition system supporting single and two-hand gestures. Uses Mediapipe for landmark detection, custom CSV logging, TensorFlow models for training, and real-time webcam gesture prediction.

A Python-based ISL/ASL Hand Gesture Recognition system supporting single and two-hand gestures. Uses Mediapipe for landmark detection, custom CSV logging, TensorFlow models for training, and real-time webcam gesture prediction.

This project uses deep learning for squat analysis, employing MediaPipe for pose detection in videos. The model classifies squat quality by analyzing key body points. After training, the model is exported to ONNX format for efficient testing and deployment across multiple platforms.

Real-time sign language detection using MediaPipe and a custom-trained neural network model. This project captures hand gestures via webcam, processes them into landmarks, and predicts the corresponding sign language gesture. It features model training with early stopping, label encoding, and one-hot encoding techniques for accurate classification.

Sign language detection uses OpenCV for real-time video capture, MediaPipe for hand/keypoint tracking, and TensorFlow for training deep learning models that translate gestures into text. This pipeline enables accurate and efficient conversion of sign language into understandable communication.

Real-time gaze-controlled robotic car using TensorFlow Lite, MediaPipe, and Raspberry Pi. Detects right-eye movements and converts them into motion commands[Close, left, right, forward] for hands-free navigation. Includes dataset capture, model training (10-fold CV), and deployment code.

A concise, end-to-end pipeline for American Sign Language (ASL) alphabet recognition in Python—covering data preprocessing with MediaPipe hand-landmark extraction, Random Forest model training and evaluation, and a real-time webcam demo that captures and “types” held gestures with audible feedback.

A Python-based app that uses Mediapipe and deep learning to detect your emotion from webcam facial and hand landmarks, then recommends YouTube music playlists matching your mood, preferred language, and favorite artist. Includes data collection, model training, and a Streamlit app for real-time emotion-based music recommendations.

Hand Gesture Classification is a Python project that uses computer vision and machine learning to classify hand gestures in real-time. By leveraging the MediaPipe framework for hand landmark detection and TensorFlow for model training, this system captures live webcam video, detects hand poses, and predicts gesture categories.

This project develop an real-time sign language detection model using technologies like Mediapipe and OpenCV. The project begins by collecting real sign language images, which serve as the training dataset for the system. The final step involves converting the recognized text into speech using the gTTS (Google Text-to-Speech) Python library.