The Google DeepDream algorithm is a modified neural network. Instead of identifying objects in an input image, it changes the image into the direction of its training data set, which produces impressive surrealistic, dream-like images. You can find the original GitHub repository at https://github.com/google/deepdream/blob/master/dream.ipynb
PsyCam is an extension of the DeepDream for the Raspberry Pi. With the RPi camera module, PsyCam can take a photo and convert it into a DeepDream.
Either follow the manual installation instructions at
http://www.knight-of-pi.org/deepdream-on-the-raspberry-pi-3-with-raspbian-jessie/
or perform the following steps for the semi-automated installation:
$ mkdir ~/deepdream && cd ~/deepdream
$ git clone https://github.com/JoBergs/PsyCam
$ cd PsyCam
$ python install_tools.py packages
$ python install_tools.py caffe
$ python install_tools.py protobuf
$ python install_tools.py camera
$ sudo reboot
The installation will take half a day or so. The installer script should also work on most modern Ubuntu systems.
The script psycam.py is controlled via command-line parameters. They are listed with
$python psycam.py --help
Start PsyCam with randomized layer and octave. This requires an attached and enabled Raspberry Pi camera module.
$python psycam.py
Start PsyCam with an input image instead of a camera snapshot (required for non-RPi usage!):
$python psycam.py -i sky_small.jpg
Make snapshots with the given size width height (large sizes will crash the RPi):
$python psycam.py -s 400 300
Make snapshots and convert them to dreams continually:
$python psycam.py -c
Start PsyCam set layer depth, type and octave manually:
$python psycam.py -d 2 -t 1 -o 6
Create a new network model file:
$python psycam.py -n
The dreams are stored in
~/deepdream/PsyCam/dreams
with the original photo and tagged with a timestamp.
Author: Johannes Bergs