Skip to content

Computational aesthetic analysis of visual media

Notifications You must be signed in to change notification settings

yilangpeng/athec

Repository files navigation

Athec

A Python package for computational aesthetic analysis of visual media

Athec is a Python library that measures a variety of aesthetic attributes, such as brightness, contrast, colorfulness, color variety, percentages of different colors, visual complexity, and depth of field. Computationally calculated visual attributes have been demonstrated to predict a wide range of outcomes, such as images' aesthetic appeal, popularity on social media, and interestingness.

How to use

  1. Install the following packages before running the scripts numpy, Pillow, matplotlib, OpenCV, Scipy, scikit-image, pyemd

The current version has been tested on the folllowing versions:

  • Python: 3.9
  • numpy: 1.20.3
  • Pillow: 8.2.0
  • matplotlib: 3.4.2
  • opencv-contrib-python: 4.5.2.54
  • scipy: 1.6.3
  • scikit-image: 0.18.1
  • pyemd: 0.5.1
  1. Run the demo scripts. The documentation about each function is also provided in these scripts.

Please note that it is recommended to download the scripts directly from this GitHub repository instead of using pip in the terminal, as the version posted there may not be actively maintained.

After downloading the scripts, please ensure that you add the package path to the sys paths as shown in the example below, or alternatively, you can revise the code in the demo scripts:

import os, sys

athec_path = os.path.expanduser("~/Documents/Workspace/Computer vision/Athec/")

sys.path.append(athec_path)

Citation

@article{peng2021athec,
  title={Athec: A Python Library for Computational Aesthetic Analysis of Visual Media in Social Science Research},
  author={Peng, Yilang},
  journal={Computational Communication Research},
  year={Forthcoming}
}

@article{peng2018feast,
  title={Feast for the Eyes: Effects of Food Perceptions and Computer Vision Features on Food Photo Popularity.},
  author={Peng, Yilang and Jemmott III, John B},
  journal={International Journal of Communication},
  volume={12},
  year={2018}
}

About

Computational aesthetic analysis of visual media

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages