Ultra-lightweight human detection. The number of parameters does not correlate to inference speed. For limited use cases, an input image resolution of 64x64 is sufficient. High-level object detection architectures such as YOLO are overkill.

ESP32 Firmware Patching Framework with Streamlined Binary/Runtime Patching Methods. Used in conjunction with Platformio and Ghidra.

In Collaboration with GDG Ranchi, Team Name: JOY BOY, Team Code: 24

Proyecto para finalización de carrera ingenieria electrónica en UTN Haedo. Se implemento un sistema que es capaz de detectar fallas por medio del espectro audible hasta los 20kHz segun las normas de SKF. Se utilizo un ESP32 para el prototipo, en conjunto con una RasberryPI 400 como servidor.

ESP32-S3 High Priority Interrupts using C

A three-tier cloud for data transfer from far-edge, to edge, to cloud, using stored images in an ESP32-CAM, to face identification on an ESP32-S3 dev, to cloud (e.g., AWS S3) and result back to the ESP32-CAM via esp32s3. Log all stats and round trip times.

Creating a smart scarecrow system to detect birds using computer vision and scare them away using directional speakers