~/talk

Edge ML with MicroPython + TinyML: Real-time Inference on under ₹500 Boards

name:Pratik
linkedin:devppratik
name:Sneha
linkedin:iam-sneha-singh
slides:none on file

Runs a fall-detector pipeline that collects MPU6050 accelerometer samples (50-sample windows), extracts magnitude mean/std and axis traces, trains a tiny TensorFlow model, then quantizes to int8 to shrink the model ~4x and convert via xxd to a C array. Deployment uses ESP32 flashed with ESP-IDF and inference with TensorFlow Lite Micro at ~1 Hz; demo shows correct fall/no-fall predictions. Notes constraints: kilobytes of RAM, power, hardware diversity, and use-triggered inference and model compression as optimizations.

AI-written summary · about

  1. 0:00Introduction and presenters
  2. 1:07Agenda and definitions
  3. 2:09Fall detection inspiration
  4. 5:46Why microcontrollers and MicroPython
  5. 8:45What is Edge ML
  6. 11:57Introduction to TinyML
  7. 14:30Step-by-step ML pipeline
  8. 21:48Live inference demo
  9. 23:15Challenges and optimizations
  10. 25:44Future possibilities and Q&A
~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.