Skip to content

Latest commit

 

History

History
81 lines (56 loc) · 3.1 KB

README.md

File metadata and controls

81 lines (56 loc) · 3.1 KB
Raw

Live Dream

Augmented reality stream video processing utilizing Google's TensorFlow + DeepDream

NOTE: This code is being actively developed. We apologize for any bugs, and thanks for reporting any issues!

Setup

Clone this repo to your local machine, and add the RNN-Tutorial directory as a system variable to your ~/.profile. Instructions given for bash shell:

git clone https://github.com/mrubash1/DeepDream_Streaming_Video
cd DeepDream_Streaming_Video
echo "export DD_STREAM=${PWD}" >> ~/.profile
echo "export PYTHONPATH=${PWD}/src:${PYTHONPATH}" >> ~/.profile
source ~/.profile

Create a Conda environment (You will need to Install Conda first)

conda create --name tf-dd-stream python=3.5
source activate tf-dd-stream
cd $DD_STREAM
pip install -r requirements.txt

Install TensorFlow

If you have a NVIDIA GPU with CUDA and cuDNN already installed:

pip install tensorflow-gpu==1.0.1

Guide for installing CUDA and cuDNN on Linux

If you will be running TensorFlow on CPU only (e.g. a MacBook Pro), use the following command (if you get an error the first time you run this command read below):

pip install --upgrade\
 https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.1-py3-none-any.whl

Error note (if you did not get an error skip this paragraph): Depending on how you installed pip and/or conda, we've seen different outcomes. If you get an error the first time, rerunning it may incorrectly show that it installs without error. Try running with pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.1-py3-none-any.whl --ignore-installed. The --ignore-installed flag tells it to reinstall the package. If that still doesn't work, please open an issue, or you can try to follow the advice here.

Download inception to models folder

cd $DD_STREAM
. build/download_inception.sh

Run unittests

We have included example unittests for the tf_train_ctc.py script

# TODO: implement some tests

Run Streaming Deep Dream Demo

All configurations for the RNN training script can be found in $DD_STREAM/configs/neural_network.ini

# TODO: implement the demo

NOTE: If you have a GPU available, the code will run faster if you set tf_device = /gpu:0 in configs/neural_network.ini

Add data

TODO: Include Data

- No data currently available

If you would like to train a performant model, you can add additional jpg files

Remove additions

We made a few additions to your .profile -- remove those additions if you want, or if you want to keep the system variables, add it to your .bash_profile by running:

echo "source ~/.profile" >> .bash_profile