Technology1 min read

Samsung running a TensorFlow lite model on 960x540 image to track you

By · Published by Everything Blog

In short

Samsung is reportedly running a TensorFlow Lite model on 960x540 images to track user behavior. The model, called "alex.tflite," is based on AlexNet and uses INT8 quantization. It compares features in 960x540 images every few seconds to determine the user's game.

Key points

  • Samsung is reportedly running a TensorFlow Lite model on 960x540 images to track user beh…: Samsung is reportedly running a TensorFlow Lite model on 960x540 images to track user behavior.
  • The model, called "alex.tflite," is based on AlexNet and uses INT8 quantization.: The model, called "alex.tflite," is based on AlexNet and uses INT8 quantization.
  • It compares features in 960x540 images every few seconds to determine the user's game.: It compares features in 960x540 images every few seconds to determine the user's game.

Sources checked by our analysis pipeline: 1

Ok, I just checked this, and its literally running a freaking TensorFlow Lite model on 960×540 images with INT8 quantization every couple of seconds in an attempt to figure out what game you are playing.

It's literally called "alex.tflite" and its literally AlexNet and seems to follow torchvision.models.alexnet. As far as I can tell this does some strange nearest neighbor feature comparisons against a DB and literally doesn't work 99% of the time, you love to see your CPU get wasted every other second by this thing though. It also just seems to dump its results into the void because the consent gate comes later.

Anyways, here's the tf model graph and some dumped captures.

Original source: twitter.com

Technology