Code for MobiCom paper 'TinyML-CAM: 80 FPS Image Recognition in 1 Kb RAM'
-
Updated
Oct 18, 2022 - Jupyter Notebook
Code for MobiCom paper 'TinyML-CAM: 80 FPS Image Recognition in 1 Kb RAM'
Pneumonia Detection on a Raspberry Pi using balenaCloud and Edge Impulse.
ROS2 wrapper for Edge Impulse
This project implements a wearable device that translates sign language gestures into text using flex sensors, an IMU, and edge ML inference.
Calido - Open Smart Thermostat and Smart Home Controller. Built on a Thingy:91 (nRF9160).
GStreamer plugin to run AI/ML inference via Edge Impulse machine learning models
This project lets you run Edge Impulse machine learning models from Rust using a safe FFI (Foreign Function Interface) layer over the Edge Impulse C++ SDK
Research on the capabilities of the STM32F429I-DISC1 board to perform Machine Learning tasks
Detector de moedas de 1 real e 50 centavos utilizando a plataforma Edge Impulse.
Submitted for NXP-Hackathon-2021 on electromaker.io
This collection of apps scrapes car images from the Autotrader website in order to create a dataset of car brands for Machine Learning (ML).
Micropython tool for Data Ingest into EdgeImpulse
A TinyML cassava leaf disease detection system built using Edge Impulse and the Espressif ESP32-CAM microcontroller
Arduino UNO R4 Minima Soccer forecast with Machine Learning (Edge Impulse)
Smart Plastic Separation Car with AI-powered detection, IoT integration, servo-controlled robotic arm, and real-time environmental monitoring.
Code example for ModusToolbox that enables any PSoC 6 to run Machine Learning
Edge-AI powered three-phase induction motor monitoring system featuring real-time electrical/vibration sensing, TinyML fault detection, predictive maintenance (EMA trend analysis), TFT UI, and a WiFi web dashboard. Built on ESP32-S3 with full open-source firmware and model pipeline.
A high-speed Serial utility for building machine learning audio datasets. This tool allows you to stream raw 16-bit PCM data from an ESP32, visualize signal levels in real-time, and save recordings into labeled .wav files.
To associate your repository with the edgeimpulse topic, visit your repo's landing page and select "manage topics."