Project: PanopticSegmentation (GitHub Link)

PanopticSegmentation-master
- instanceSegmentation.py
- createPanopticResultsImageFile.py
- config.yaml
- installInstanceSegWeights.py
- classConversionDict.py
- LICENSE
- dependencies
  - Mask_RCNN
    - samples
      - balloon
        README.md
        balloon.py
      - nucleus
        README.md
        nucleus.py
      - shapes
        train_shapes.ipynb
        shapes.py
      - coco
        coco.py
    - images
    - LICENSE
    - setup.py
    - setup.cfg
    - README.md
    - mrcnn
      - config.py
      - parallel_model.py
      - model.py
      - __init__.py
      - utils.py
      - visualize.py
    - requirements.txt
    - .gitignore
    - MANIFEST.in
    - assets
  - deeplab-pytorch
    - eval.py
    - libs
      - datasets
        cocostuff.py
        __init__.py
      - utils
        crf.py
        metric.py
        loss.py
        __init__.py
      - models
        deeplabv2.py
        deeplabv3plus.py
        resnet.py
        __init__.py
        deeplabv3.py
        msc.py
      - __init__.py
      - caffe.proto
    - convert.py
    - LICENSE
    - demo.py
    - config
      - cocostuff10k.yaml
      - voc12.yaml
      - README.md
      - cocostuff164k.yaml
      - conda_env.yaml
    - data
      - datasets
        voc12
        labels.txt
        cocostuff
        labels_2.txt
        labels.txt
        cocostuff_hierarchy.yaml
      - models
        deeplab_resnet101
        init
        .gitkeep
        voc12
        .gitkeep
        coco_init
        .gitkeep
        cocostuff10k
        .gitkeep
        cocostuff164k
        .gitkeep
        .gitignore
    - train.py
    - README.md
    - scripts
      - demo_default.sh
      - setup_cocostuff164k.sh
      - setup_cocostuff10k.sh
      - setup_caffemodels.sh
      - eval_default.sh
    - .gitignore
    - docs
    - hubconf.py
    - livedemo.py
  - panopticapi
    - license.txt
    - cityscapes_gt_converter
      - cityscapes_panoptic_converter.py
    - panoptic_coco_categories.json
    - combine_semantic_and_instance_predictions.py
    - evaluation.py
    - converters
      - panoptic2semantic_segmentation.py
      - 2channels2panoptic_coco_format.py
      - __init__.py
      - detection2panoptic_coco_format.py
      - panoptic2detection_coco_format.py
    - sample_data
      - panoptic_coco_detection_format.json
      - panoptic_examples_2ch_format
      - images_info_examples.json
      - panoptic_examples.json
      - input_images
      - panoptic_examples
      - panoptic_examples_testsub_results
        panoptic_examples_testsub_results.json
        panoptic_examples_testsub_results
      - panoptic_examples_testsub_results.zip
    - converted_data
      - .gitignore
    - README.md
    - utils.py
    - visualization.py
    - CONVERTERS.md
- semanticSegmentation.py
- Pipfile
- Pipfile.lock
- panopticSegmentation.py
- __init__.py
- README.md
- scripts
  - download_dataset.sh
  - menu.sh
  - download_and_unpack_results.sh
  - clone_to_zip.sh
  - raw_repo_setup.sh
  - eval_panoptic_results.sh
  - zip_files.sh
  - deeplab_setup.sh
  - run_inference_and_zip.sh
  - download_and_eval_results.sh
- .gitignore
- static
- helpers.py

PanopticSegmentation

This Repository combines the MASK_RCNN algorithm for instance segmentation and DeepLabV2 Algorithm for semantic segmentation in order to produce predictions for the Panoptic Segmentation Challenge.

Alt text

In order to use, it's require to setup the repository first:

Prerequisites

Make sure you have Python 3.6 and pip3 9.0.1 and pipenv (other pip versions may work, however they are untested) This also requires CUDA 9 and cuDNN to be installed on system.

Installation:

git clone to clone the repository
Enter Root directory:
```
cd PanopticSegmentation
```
Install dependencies:
```
pipenv install
```

Usage

From inside the root directory:

Enter Virtual Env:

pipenv shell

Run the command:

pipenv run start

Follow the instructions in the menu to run either inference or evaluation of the dataset.