Skip to content

Simple computer vision mechanism that given the 10 most prominent colors of a given image, tries to classify from which time period it comes. This classification is based on learning the color pattern of thousands of publicly available artpieces.

License

Notifications You must be signed in to change notification settings

vieira-giulia/ColorCorrelationInArtHistory

Repository files navigation

Color Correlation In Art History

what can the colors of a painting tell about it's time context? a computer vision approach

Universidade Federal de Minas Gerais BSc. Computer Science 2020.1 subject: Computer Vision Giulia Vieira

Files description

Paper | paper.pdf


Code | code.ipynb


Presentations | presentation01.pdf https://youtu.be/shwqzlfmKRQ | presentation 02.pdf https://youtu.be/5TPAEhZofSk | presentation 03.pdf https://youtu.be/qRj03hEcpq8


Image-Color from SemArt | image_colors.csv


ImageName-Date from SemArt| name_date.csv


RGB-HEX translator | colors.csv

References:

METHOD https://towardsdatascience.com/building-a-color-recognizer-in-python-4783dfc72456

INSPIRATION https://artsexperiments.withgoogle.com/artpalette/colors/261a27-322630-816438-ca9950-4c3731

DATASET SemArt: http://researchdata.aston.ac.uk/id/eprint/380/ ArtImage: https://www.kaggle.com/thedownhill/art-images-drawings-painting-sculpture-engraving

COLORS https://github.com/codebrainz/color-names/blob/master/output/colors.csv

About

Simple computer vision mechanism that given the 10 most prominent colors of a given image, tries to classify from which time period it comes. This classification is based on learning the color pattern of thousands of publicly available artpieces.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published