Skip to content

An extension of Mask RCNN model for training on custom data or masking videos rather than images

Notifications You must be signed in to change notification settings

piy0999/MaskRCNN-Extension

Repository files navigation

Step 1 - Create a Conda virtual environment

We will be using Anaconda with python 3.6. If you don't have Anaconda, follow this tutorial https://www.youtube.com/watch?v=T8wK5loXkXg

Create a anaconda virtual environment by:

conda create -n MaskRCNN anaconda python=3.6

Enable environment support in jupyter:

conda install nb_conda

Step 2 - Install the Dependencies

Activate the virtual environment, install dependencies and run setup:

conda activate MaskRCNN
conda install jupyter
pip3 install -r requirements.txt
python3 setup.py install

Step 3 - Install pycocotools

Needed to use Microsoft CoCo dataset:

pip3 install git+https://github.com/waleedka/coco.git#subdirectory=PythonAPI

Step 4 - Launch Jupyter

jupyter notebook

Step 5 - Open MaskRCNN Notebook

Everything is in MaskRCNN.ipynb.

Note: Video processing is in Video_Masker.ipynb.

About

An extension of Mask RCNN model for training on custom data or masking videos rather than images

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published