Skip to main content

A full-reference quality metric for analyzing restoration methods.

Project description

ERQA - Edge Restoration Quality Assessment

ERQA - a full-reference quality metric designed to analyze how good image and video restoration methods (SR, deblurring, denoising, etc) are restoring real details.

It is part of MSU Video Super Resolution Benchmark project.

Quick start

Run pip install erqa and run it from command line or directly from Python code.

Command line

python -m erqa /path/to/target.png /path/to/gt.png

Python code

import erqa
import cv2

# Target and gt should be uint8 arrays of equal shape (H, W, 3) in BGR format
target = cv2.imread('/path/to/target.png')
gt = cv2.imread('/path/to/gt.png')

metric = erqa.ERQA()
v = metric(target, gt)

Description

The ERQA metric analyzes how details were reconstructed in an image compared to ground-truth.

  • ERQA = 1.0 means perfect restoration
  • ERQA = 0.0 means worst restoration

Visualization of the metric shows underlying mask showing where image is distorted.

  • Blue means there is a missing detail (False Negative)
  • Red means there is a misplaced detail (False Positive)
  • White means perfect details restoration (True Positive)
  • Black means perfect background restoration (True Negative)

Local setup

You can get source code up and running using following commands:

git clone https://github.com/msu-video-group/erqa
cd erqa
pip install -r requirements.txt

Cite us

Soon

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

erqa-1.1.2.tar.gz (4.8 kB view details)

Uploaded Source

Built Distribution

erqa-1.1.2-py3-none-any.whl (5.4 kB view details)

Uploaded Python 3

File details

Details for the file erqa-1.1.2.tar.gz.

File metadata

  • Download URL: erqa-1.1.2.tar.gz
  • Upload date:
  • Size: 4.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.60.0 CPython/3.9.6

File hashes

Hashes for erqa-1.1.2.tar.gz
Algorithm Hash digest
SHA256 86a72d9a26c59842c52e47248de8ef09e11b688a009359f6182dc30a55cea81b
MD5 560f1aeab7d353a77bfb6759d6e4ebec
BLAKE2b-256 6a0bac8864bd05a22ed66d90efecea08857f2664bd0598c1cd9c4f77882840ba

See more details on using hashes here.

File details

Details for the file erqa-1.1.2-py3-none-any.whl.

File metadata

  • Download URL: erqa-1.1.2-py3-none-any.whl
  • Upload date:
  • Size: 5.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.60.0 CPython/3.9.6

File hashes

Hashes for erqa-1.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 32c454fefe28467ab58d97ca005e45a46fabca554b44c0004467508d8a5486d4
MD5 591d2ed106c38e351f4ed152e15ce4e7
BLAKE2b-256 c982f91db221dc86c6b510e636b18154d782c1f87b1693bcca860d43a376615c

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page