tensorflow-lite

Visualizer for neural network, deep learning and machine learning models

Repository for the "Machine Learning for Physical Computing" class at ITP, NYU

Explore the fascinating world of depth perception with this cutting-edge web application! Using the power of TensorFlow.js and the MiDaS (Monocular Depth Estimation) model converted to TensorFlow Lite format, this project brings real-time depth prediction right to your browser.

Device-side machine learning-based pull-up counter using ml5.js, a TensorFlow.js-based machine learning library.

MoonMakers-Pinball in a game that works in the browser, where we use our Arduino Nano 33 BLE Sense, as a command to control the imposition of our ball thanks to AI.

A simple image classification app created using tensorflow-js and react. The app uses the pre-built MobileNet v2 model for image classification without going through any additional training process.

Sample robot apps for Kaia.ai platform and Scout robot base

An app that classifies a photo taken on mobile devices or a selected image of a cat or dog.

Bird! Bird! Bird! A RaspberryPi W2 Bird Watcher project

Real-time background removal using a webcam feed, pure JavaScript, and the TensorFlow Lite (TFLite) library for portrait segmentation.

Fastron - Optimized Neural Network Visualizer

Secured Insurance Data - hackzurich weekend hack

AI-Powered Smart Agriculture Platform that detects crop diseases using Deep Learning (CNN) and TensorFlow Lite, providing real-time disease diagnosis, crop health monitoring, and intelligent farming recommendations through a React Native mobile application backed by FastAPI and Firebase.

Find personal data on the device, send the LLM placeholders, restore the reply locally. ONNX Runtime Web in the browser, LiteRT on Android.

Yogasana assists users in achieving optimal posture during yoga practice through real-time pose detection.

Accessible AAC communication app built in React — text, voice, pictures & sign language input, wrapped in a gamified "garden" that grows as you communicate. On-device MediaPipe + TFLite vision pipeline for sign tracking.

Filipino Sign Language assistant for real-time text and speech on the web, Android, and desktop.

Chest X-Ray reports in a matter of seconds.We have created a secure Chest X-Ray image classification based website and app that can detect Covid-19, Pneumonia and Tuberculosis. Deep Learning has been used to detect the disease by using a Convolutional Neural Network(MobileNetV2) which performs classification

AI-powered resistor colour detection and resistance value prediction using deep learning.