3.15 Demonstration of computer vision system “thumbs up”

The demonstration of how to use a pre-trained Resnet-18 neural net to detect 👍 or 👎 in a video stream using Nvidia’s Jetson Nano is based on the Nvidia course Getting Started with AI on Jetson Nano

The Jetson Nano is a small and powerful computer for a wide range of ML applications

The choice on courses is small and ML centered. All course work is in Python and is run in hosted notebooks.

ML applications for Jetson Nano:

  • image classification
  • object detection
  • segmentation
  • speech processing

Despite the low price of €99 and the size of 69 mm x 45 mm, the technical specification is quite impressive

Jetson Nano technical specification:

  • GPU: 128-core Maxwell
  • CPU Quad-core ARM A57 @ 1.43 GHz
  • Memory 4 GB 64-bit LPDDR4 25.6 GB/s
  • Camera 2x MIPI CSI-2 DPHY lanes

3.15.1 Run demonstration on Jetson Nano

To run the demonstration ssh into the Jetson Nano using its IP address, start jupyter notebook which opens a window in the browser from which you can select the demonstration.

ssh dlinano@192.x.x.x
jupyter notebook

# open in browser on port 8888
http://192.x.x.x:8888/lab?