Art Scene Disrupting the Market Through Inclusion

Art Scene (formerly AI Art Advisor), the next-generation art discovery and evaluation app, is disrupting the art market by combining cutting-edge technology with deep insights into art and aesthetics. Our proprietary "artistic quotient" machine learning system helps users discover their unique taste in art and navigate the art market with confidence. With Art Scene, collecting art is no longer limited to the elite few - our app is democratizing the market and making it accessible to everyone.

post-template-default,single,single-post,postid-4203,single-format-standard,wp-custom-logo,qi-blocks-1.2.7,qodef-gutenberg--no-touch,stockholm-core-2.4,qodef-qi--no-touch,qi-addons-for-elementor-1.7.1,select-theme-ver-9.9,ajax_fade,page_not_loaded,side_area_over_content,,qode_menu_,qode-mobile-logo-set,elementor-default,elementor-kit-550,elementor-page elementor-page-4203

Paintings Over Time

Visualizing Color Usage

Paintings Over Time

People have been fascinated by color palettes throughout history, including historians, archaeologists, and art lovers. As the famous Russian painter Wassily Kandinsky once wrote: “Color is a power which directly influences the soul.”

The use of color in art varies from culture to culture, season to season. Using an interactive visualization, we aim to examine color usage in artworks over time across regions. We will proceed in the following manner with this project:

  1. Review the literature.
  2. Brainstorm ideas and draft the interface for the visualization
  3. Identify and refine datasets
  4. Make a mockup visualization using a small set of data
  5. Improve visualization by adding enhanced features
  6. Utilize a large museum art dataset

For the major progress checkpoints, we plan to accomplish the following:

  • Mid-term: a rudimentary interactive visualization showing dominant and
    secondary colors over time across different regions.
  • Final: enhanced visualization, utilizing larger museum art data with additional functionality.


Literature Review


Color Theories

There are many theories about color. Ancient Greeks believed colors emerged from light-dark struggles. Heraldry developed rapidly during the late Middle Ages, often using tinctures. Color meaning is affected by mixing pigments, according to Leonardo da Vinci (1452-1519). In the 1600s, color was used more liberally. The use of pigment has no longer been restricted to specific purposes. Sunlight was split by Isaac Newton (1642-1727). He founded color theory in his Optics 1 in 1704. Johann von Wolfgang Goethe (1749–1832) in 2 (1791) studied how colors affected emotions and psychology. His six-hued spectrum continues to inspire artists today. A three-dimensional color system based on hue, value, and saturation was proposed by Albert Munsell (1858–1928) 3. Otto Philip Runge (1770–1840) founded complementary color theory in 1810. Two innovations in art emerged from the Industrial Revolution: synthetic pigments and tube paint. By the time pop art arose, we expected that any mass-produced product would be available in any color we wanted. Warhol’s Marilyn appears equally valid in all its color variations. It doesn’t matter what color is used, as long as it’s deliberate. Today, digital tools make it possible for us to see color as an infinite space.

Art Painting Image Color Analysis

Colors influence each other. There are a lot of researchers working in this area, including color-classifying search engines 4, color-based art movement classification 5, image retrieval by color semantics 6, and semantic issues in art painting 7.

Download the App

Inspired Me

“The app inspired an artistic part of me that’s always been there but hasn’t been tapped into for quite some time.”

Nini Amerlise

Winner of Supermodel Canada

Challenges Snobbish Ideals

“This innovative use of technology challenges the snobbish idea that only the rich can afford great art by helping anyone learn how to confidently and affordably bring the power of beauty into their lives.”

Sean Latham

Zarrow Center Director

Discover Yourself

“It’s fun. It helps you discover something about yourself and gives you an idea of what speaks to you.”

Nicole Mölders

Alaska, US

Color Quantization

By simplifying the number of colors in an image, Color Quantization is often used to determine the dominant and secondary colors. Algorithms such as Median Cu, K-means, and Octreesm are typical. Fine art computer systems illustrate the various features they support. A few examples follow: image retrieval system for paintings8,art historian system9, PICASSO system10, painting classification system11, and etc.


Besides the Budget Collector’s dataset, we looked at data from the Art Institute of Chicago, Rijksmuseum, Museum of Modern Art (MoMA), and The Met.

  • Through its Open API, The Art Institute of Chicago offers a variety of search endpoints, including the collection, artists, artworks, and exhibitions.
  • Rijksmuseum provides access to object metadata, bibliographic data, lexicon, as well as user-generated content.
  • With almost 200,000 artworks, MoMA has one of the largest art collections in the country. Although they don’t have a public API, their collection/artist data can be downloaded as .csv or .json files.
  • In 2017, The Met launched its Collection API to make public domain artwork accessible to anyone. Data was available as a .csv download on GitHub or can be accessed directly through their API.


Considering museum data usually does not include images readily available, we will first create our mock-up visualization using the Budget Collector dataset, which contains hundreds of pieces of artwork ranging from the 13th century to the 21st century. Each record includes an image and a more extensive selection of numerous fields.

Our next phase of the project will involve the use of larger museum data. At the moment, we plan on using the MoMA dataset. There are 98,361 artworks on the MoMA website created by 27,140 artists. A total of 140,848 records are included in this MoMA dataset, representing all of the works that are now part of MoMA’s collection. We will look at the following datasets, joining them, extracting needed fields, creating a sample, and extracting images associated with the sampled records.

  • There are 140,848 records in the Artworks dataset, representing all the works in MoMA’s collection and cataloged in the database. Each work includes basic metadata, such as title, artist, medium, dimensions, and date acquired by the museum.
  • The Artists dataset contains 15,243 records, representing all the artists whose work is in MoMA’s collection and is cataloged in the database. The basic metadata include name, nationality, gender, birth year, death year, Wiki QID, and Getty ULAN ID for each artist. The MoMA flags incomplete records as “not curator approved.” We will exclude these records.



  1. 1. Newton I. Optics. Dover, New York, 1952
  2. Goethe J. Goethe’s Color Theory. New York, Van Nostrand Reinhold Company, 1971
  3. Munsell A. A Grammar of Colors. New York, Van Nostrand Reinhold Company, 1969
  4. Gevers T., A.W.M. Smeulders. Image Search Engines: An Overview. University of Amsterdam, June 2003
  5. G ̈unsel B., S. Sariel, O. Icoglu. Content-Based Access to Art Paint-ings. The IEEE International Conference on Image Processing (ICIP), 2005.
  6. Corridoni J. M., A. Del Bimbo, P. Pala. Image Retrieval by ColorSemantics. Multimedia Systems, 7, 3, May, 1999, 175–183.
  7. Stanchev P., D. Green Jr., B. Dimitrov. Some Issues in the Art Image Database Systems. Journal of Digital Information Management, Vol.4, Issue 4 (December 2006), 227–232.
  8. Lombardi T., C. Sung-Hyuk, Ch. Tappert. A Lightweight Image Re-trieval System for Paintings. Volume Storage and Retrieval Methods and Applications for Multimedia, 5682, (2005), 236–246.
  9. Icoglu O., G. Bilge, S. Sanem. Classification and Indexing of Paintings Based on Art Movements. 12th European Signal Processing Conference(Eusipco) 2004, September 6–10 2004, Vienna, Austria, 749–752.
  10. Del Bimbo A., P. Pala. Visual Querying by Color Perceptive Regions. Fourth DELOS Workshop Image Indexing and Retrieval, San Miniato,1997, 2–6.
  11. Keren D. Painter Identification Using Local Features and Na ̈ıve Bayes. Proceedings of the 16th International Conference on Pattern Recognition,Quebec, Canada, 2002, Vol. 2, 474–477.