DiaboloKiat/project-Mask-RCNN

project-Mask-RCNN (Machine Learning)

Mask R-CNN

• IEEE
• paper
• github

Backbone Network Architectures

• ResNeSt 2020
  • paper
  • ResNeSt | Pytorch
  • github
  • detectron2-ResNeSt
• ResNeXt 2017
  • paper
  • ResNeXt | Pytorch
  • github
• ResNet 2015
  • paper
  • ResNet | Pytorch
  • github

Dataset

• Cocodataset
• How to download
  • !wget http://images.cocodataset.org/zips/train2017.zip (Using in Colab)
  • !wget http://images.cocodataset.org/zips/val2017.zip (Using in Colab)
  • !wget http://images.cocodataset.org/annotations/annotations_trainval2017.zip (Using in Colab)

---

How to clone repo

If you have ssh-key, you can run this command

$ git clone git@github.com:DiaboloKiat/project-Mask-RCNN.git

If you have not ssh-key, you can run this command

$ git clone https://github.com/DiaboloKiat/project-Mask-RCNN.git

---

How to run docker

First you need to build docker

$ cd ~/project-Mask-RCNN/docker
$ source docker_build.sh

Now you can run your docker

$ cd ~/project-Mask-RCNN/docker
$ source docker_run.sh

If you want to open a second terminal with docker, you can run this command

$ cd ~/project-Mask-RCNN/docker
$ source docker_join.sh

---

How to open jupyter notebook

After you go into the docker container, you can run this command to open jupyter notebook

$ jupyter notebook

• Then press the ctrl key and click on the displayed URL to enter jupyter notebook
• All the .ipynb file will at the ～/project-Mask-RCNN/jupyter/, there are three .ipynb file, which are respectively Mask_RCNN_ResNet, Mask_RCNN_ResNeXt and Mask_RCNN_ResNeSt.