Computer Image Likeness Assessing Automation

Overview

CILISSA allows for the use of various metrics to perform full-reference image comparisons.

It features the most popular full-reference image quality metrics, image transformations and translations. CILISSA is also very extensible and new operations can be easily added.

CILISSA has an optional Qt-based graphical interface that lets you experiment with various operations, their orders and properties.

Requirements

• Python >=3.9,<3.12

Installation

Install from PyPI

$ pip install cilissa

Releases

Binaries for Windows and Linux can be found on GitHub releases.

Usage

GUI

Information about the GUI can be found in the cilissa_gui/README.md file.

CLI

Currently the CLI only supports working with a single pair of images.

The parameters of metrics and transformations can be modified by passing them to the --kwargs argument using the following format:

<operation-name>-<parameter-name>=<value>

where parameter-name uses hyphens (-) instead of underscores (_)

Documentation

Documentation is hosted on Read the Docs.

License

CILISSA is under the terms of the MIT License, following all clarifications stated in the license file.