The videos used for generating the dataset are available here.
The documentation is available here.
$ git clone <link_to_forked_repository>
$ cd /path/to/Memoir/
For Python (virtual) environment:
$ pip3 install numpy
$ pip3 install scikit-image
$ pip3 install opencv-python
$ pip3 install tensorflow==1.14 (cpu)
$ pip3 install tensorflow-gpu==1.14 (gpu)**
** You will need CUDA and cuDNN libraries to run tensorflow on GPU.
Building Tensorflow 1.14 with GPU Support and TensorRT on Ubuntu 16.04 LTS
Building Tensorflow 2.0 with GPU support and TensorRT on Ubuntu 18.04 LTS
For Conda environment:
Either use anaconda-navigator
to install the dependencies, including tensorflow-gpu==1.14. It will automatically take care of CUDA and cuDNN. Or you could just duplicate the developer's environment using:
conda env create -f requirements.yml
For both the environments.
(Memoir)$ pip install .
to install in editable (developer) mode:
(Memoir)$ pip install -e .
$ pip uninstall -y Memoir