Published July 30, 2024 | Version v1
Dataset Open

Deep learning based Image Compression for Microscopy Images: An Empirical Study

  • 1. ROR icon Leibniz Institute for Analytical Sciences - ISAS

Contributors

Supervisor:

  • 1. ROR icon Leibniz Institute for Analytical Sciences - ISAS
  • 2. ROR icon Ruhr University Bochum

Description

This dataset supports the manuscript titled "Deep Learning Based Image Compression for Microscopy Images: An Empirical Study." The materials provided are essential for replicating the experiments in the study. The dataset is divided into three main parts:

  1. Config: train and test yaml config files for both 2D and 3D mmv_im2im labelfree task
  2. Sample_data: sample data to test the compression and the downstream labelfree code, in both 2D and 3D case.
  3. Model: fine-tuned 2D/3D labelfree model and 3D compression model. We use pre-trained 2D compression models from CompressAI

 

Files

config.zip

Files (2.1 GB)

Name Size Download all
md5:556d62fac3b66dbf16e84a4d79d0d7a1
68.3 kB Preview Download
md5:2046b67aa2a6603c071dcd132c8b527b
1.9 GB Preview Download
md5:074459e7ea3497215e154edff79be9b3
259.0 MB Preview Download

Additional details

Software

Repository URL
https://github.com/MMV-Lab/data-compression
Programming language
Python
Development Status
Wip