https://github.com/volume-em/empanada

Panoptic segmentation algorithms for 2D and 3D electron microscopy images
https://github.com/volume-em/empanada

deep-learning electron-microscopy instance-segmentation panoptic-segmentation pytorch segmentation semantic-segmentation

Last synced: about 2 months ago
JSON representation

Panoptic segmentation algorithms for 2D and 3D electron microscopy images

• Host: GitHub
• URL: https://github.com/volume-em/empanada
• Owner: volume-em
• License: gpl-2.0
• Created: 2021-12-01T16:02:54.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2023-02-25T22:56:09.000Z (over 1 year ago)
• Last Synced: 2024-06-20T05:49:33.629Z (3 months ago)
• Topics: deep-learning, electron-microscopy, instance-segmentation, panoptic-segmentation, pytorch, segmentation, semantic-segmentation
• Language: Python
• Homepage: https://empanada.readthedocs.io/en/latest/index.html
• Size: 25.6 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 6
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-biological-image-analysis - Empanada - Panoptic segmentation algorithms for 2D and 3D electron microscopy images. (Electron and super resolution microscopy)

README

        
# empanada

## Overview

empanada (**EM** **Pan**optic **A**ny **D**imension **A**nnotation) is a tool for panoptic segmentation of organelles in 2D and 3D electron microscopy (EM) images.

Panoptic segmentation combines both instance and semantic segmentation enabling a holistic approach to image annotation

with a single deep learning model. To make model training and inference lightweight and broadly applicable to many EM

imaging modalities, empanada runs all expensive operations on 2D images or run length encoded versions of 3D volumes.

**Note: Development is active and breaking changes should be expected. Not all features are implemented.**

## Highlights

  - Train 2D panoptic segmentation models by customizing easy to read .yaml config files.

  - Get better models faster by using state-of-the-art CEM pretrained model weights to initialize training.

  - Run models on 2D images, videos, or isotropic and anisotropic volumes with orthoplane and stack inference, respectively.

  - Export GPU and quantized CPU trained models for use in napari with [empanada-napari](https://github.com/volume-em/empanada-napari).

## Installation

Install empanada with pip:

```shell

pip install empanada-dl

```

To install the latest development version of empanada directly from GitHub:

```shell

pip install git+https://github.com/volume-em/empanada.git

```