Demo Abstract: Image Storage and Broadcast over BLE with Deep Neural Network Autoencoding

Chong Shao, Shahriar Nirjon

Abstract: This demo in an implementation of a new Deep Image Beacon system that is capable of broadcasting color images over a very long period (years, as opposed to days or weeks) using a set of cheap, low-power, memory-constrained Bluetooth Low Energy (BLE) beacon devices. We adopt a deep neural network image encoder to encode the given input image and generates a compact representation of the image. The representation can be as short as 10 bytes. On the receiver end, we adopt a deep neural network decoder running on a mobile device. When the mobile device receives the BLE broadcasted image data, it decodes the original image. We develop a pair of smartphone applications. One application takes an image and user-requirements as inputs, shows previews of different quality output images, writes the encoded image into a set of beacons. The second application reads the broadcasted image back.